Now you can get help with all of this and more by simply asking an AI agent to take care of it — while you drink a second cup of coffee and focus on your team’s long-term strategy.

An agent can tackle certain tasks with you or for you, from acting as a virtual project manager to handling more complex assignments like reconciling financial statements to close the books. Microsoft 365 Copilot is already a personal assistant that helps with everything from tedious daily duties to jumpstarting creative projects. Using it to interact with various agents brings a new world of possibilities for organizations to empower their employees, drive business and accomplish even more.

Agents can operate around the clock to review and approve customer returns or go over shipping invoices to help businesses avoid costly supply-chain errors. They can reason over reams of product information to give field technicians step-by-step instructions or use context and memory to open and close tickets for an IT help desk.

“Think of agents as the new apps for an AI-powered world,” says Jared Spataro, Microsoft’s chief marketing officer for AI at Work. “We’re rapidly adding new capabilities to tackle individuals’ biggest pain points at work and drive real business results.”

What are agents, anyway?

An agent takes the power of generative AI a step further, because instead of just assisting you, agents can work alongside you or even on your behalf. Agents can do a range of things, from responding to questions to more complicated or multistep assignments. What sets them apart from a personal assistant is that they can be tailored to have a particular expertise.

For example, you could create an agent to know everything about your company’s product catalog so it can draft detailed responses to customer questions or automatically compile product details for an upcoming presentation.

Other agents can do even more, acting on your behalf, like one that helps fulfill sales orders — freeing you up to focus on building new customer relationships. Having agents handle some of these routine needs can boost productivity across industries, from manufacturing and research to finance and retail, helping businesses save time and money.

You can use ready-made agents in Microsoft 365 and Dynamics 365, or build custom agents to help with more specific needs in Copilot Studio.

Imagine you’re a salesperson with big quarterly goals to meet. Copilot acts as your personal assistant, drafting emails, recapping a meeting you missed and helping you design a polished sales presentation. Meanwhile, an agent specialized in sales lead generation works autonomously in the background to find new prospects you can follow up with later in the week. Copilot partners on daily tasks, and your purpose-built agent uses its customized skills to help you meet your end-of-quarter goals.

AI agents are not only a way to get more value for people but are going to be a paradigm shift in terms of how work gets done.

Agents are not new. Microsoft has done extensive research in the area and even created a multi-agent library last year for developers around the world, work that helped shape what agents can do today.  They’re getting more attention now because recent advances in large language models (LLMs) help anyone — even outside the developer community — communicate with AI. That agent-LLM duo makes AI tools more tangibly useful.

“People expect AI to do things for them,” not to just generate language, says Ece Kamar, the managing director of Microsoft’s AI Frontiers Lab. “If you want to have a system that can really solve real world problems and help people, that system has to have a good understanding of the world we live in, and when something happens, that system has to perceive that change and take action accordingly.”

Agents are like layers on top of the language models that observe and collect information, provide input to the model and together generate an action plan and communicate that to the user — or even act on their own, if permitted. So both agents and models are equally important pieces of the puzzle, as far as generative AI tools go.

Agents will become more useful and able to have more autonomy with innovations in their three necessary elements: memory, entitlements and tools.

Memory helps provide continuity so that each time you ask for something, it isn’t like starting from scratch.

“To be autonomous you have to carry context through a bunch of actions, but the models are very disconnected and don’t have continuity the way we do, so every prompt is in a vacuum and it might pull the wrong memory out,” says Sam Schillace, Microsoft’s deputy chief technology officer. “It’s like you’re watching a stop-motion animation, one isolated frame after another, and your mind puts it into motion. The clay model doesn’t move on its own.”

To build up the memory infrastructure to address this, Schillace and his team are working on a process of chunking and chaining. That’s essentially what it sounds like: They’re experimenting with dividing up interactions in bits that can be stored and linked together by relevance for faster access, akin to a memory — like grouping conversations about a certain project so an agent can recall those details when you ask for a status update and not have to search through its entire database.

The work with entitlements and tools is making sure agents have secure access to, or are entitled to, information they need in order to accomplish things for you, with your permission — like who your boss is, for example — and to the computer programs they need to take action on your behalf, like Teams and PowerPoint.

How to use and build agents for work

You can already create and publish agents in Microsoft 365 Copilot that can help you in your daily work as easily as you’d create a spreadsheet or presentation — no coding skills required.

You don’t need to be a developer to build agents using Copilot Studio, either. Anyone can connect them to relevant business data such as emails, reports and customer management systems so they can perform tasks and provide insights.

And you’ll soon be able to enlist new agents in Microsoft 365 to help with common workflows and tasks. Interpreter in Teams will provide real-time speech-to-speech translation during meetings, for example, and you can opt to have it simulate your voice. The Employee Self-Service Agent will simplify human resource and IT help desk-related tasks like helping workers resolve a laptop issue or find out if they’ve maxed out certain benefits, and it can connect to company systems for further customization in Copilot Studio.

Microsoft Dynamics 365 will have agents as well for a range of common business workflows across sales, supply chain, finance and customer service functions.

And every SharePoint site will soon come equipped with an agent tailored to your organization’s content that allows employees to quickly tap into these vast knowledge bases and find exactly what they need in seconds, whether it’s project details buried in a workback schedule or a summary of a recent product memo.

Developers have even more options. With the new Azure AI Agent Service, you’ll be able to choose from small or large language models to orchestrate, develop and scale agent-powered apps to streamline and automate complex workflows like order processing and customer data synchronization. It provides a software development kit with tools for developing agents, allowing you to efficiently integrate agent capabilities using Visual Studio Code and GitHub.

One type of model, OpenAI’s recently announced o1 series, can bring more advanced reasoning capabilities to agents, allowing them to take on more complicated tasks by breaking them down into steps — like getting the information someone on an IT help desk would need to solve a problem, factoring in solutions they’ve tried and coming up with a plan.

You can also use the power of agents in LinkedIn; the platform’s first agent can help recruiters with hiring.

Assessing risk for autonomous action

There are extra safety considerations with agents that can act autonomously, and Microsoft is focused on making sure agents only access what you want them to, says Sarah Bird, the company’s chief product officer of Responsible AI.

“Agents certainly up the stakes from a responsible AI point of view,” Bird says. “So we have to have much, much lower error rates. And there’s many more nuanced ways in which something could be an error. This is the big challenge with agents.”

But the same responsible AI foundational playbook for other AI applications can be used to assess and mitigate risk with agents, she says.

The new Copilot Control System helps IT departments manage Copilot and agents with data access and governance, management and security controls, as well as measurement reports and tools to track adoption and business value.

Many agents, like those created for Microsoft 365 and Dynamics 365, include “human in the loop” approvals, where people are required to take the final step of reviewing and sending an email the Sales Order Agent wrote, for example. And for agents developed in Copilot Studio, authors can review the records to see which actions the agent took and why.

The key is to focus on testing and moderating to ensure accuracy, Bird says, and for organizations to choose the right starting point for their needs.

“We will of course make progress by building on the foundation we already have, so we’re starting the journey from a strong place,” Bird says.

Looking back — and into the future

Technologists have long been excited by the idea of autonomous systems working side-by-side with people to help them, says Kamar, who has been working on AI agents since 2005 and even wrote her Ph.D. thesis on the topic in 2010. The hurdle was that “we lacked that general problem-solving power” on the back end, she says.

With LLMs, “we finally have this missing component,” she says. “Now we can bring back a lot of the ideas from our decades of research.”

Going forward, Kamar envisions a new ecosystem or marketplace of agents, sort of like how apps empower people to do more with their smartphones.

Agents already have “the basic building blocks of what it takes to complete a task,” she says. “Like observing, ‘I can see your meeting is taking longer; I should delay the next meeting.’”

They’re getting more helpful as they gain autonomy through the innovations in memory and entitlements. They’re relieving pain points for employees by helping with things like expense reporting, project management and meeting facilitation. And they’re driving exponential impact for businesses by taking on duties like alerting supply chain managers to low inventory and then automatically reordering to help drive sales and keep customers satisfied.

Agents matter because they “open up a whole set of opportunities for working with people for getting tasks done, and that’s what we expect from AI systems,” Kamar says. “AI agents are not only a way to get more value for people but are going to be a paradigm shift in terms of how work gets done.”

And this is just the beginning. Copilot is set to evolve with new capabilities like Copilot Actions, designed to handle routine tasks that can bog down employees like summarizing emails missed during time off, compiling agenda items and generating monthly reports. More capabilities like these are coming over the next year to lift the weight of work for employees and teams.

“Copilot will empower every employee to do their best work in less time, and focus on more meaningful tasks,” Spataro says. “And agents created in Copilot Studio will transform every business process, helping companies streamline operations, enhance collaboration and drive innovation at scale.”

Illustrations by Michał Bednarski / Makeshift Studios

The post AI agents — what they are, and how they’ll change the way we work appeared first on The Microsoft Cloud Blog.

One of the highlights of my career has always been connecting with customers and partners across industries to learn how they are using technology to drive their businesses forward. In the past 30 years, we’ve seen four major platform shifts, from client server to internet and the web to mobile and cloud to now—the next major platform shift to AI.

As today’s platform shift to AI continues to gain momentum, Microsoft is working to understand just how organizations can drive lasting business value. We recently commissioned a study with IDC, The Business Opportunity of AI, to uncover new insights around business value and help guide organizations on their journey of AI transformation. The study found that for every $1 organizations invest in generative AI, they’re realizing an average of 3.7x return—and uncovered insights about the future potential of AI to reshape business processes and drive change across industries. 

Today, more than 85% of the Fortune 500 are using Microsoft AI solutions to shape their future. In working with organizations large and small, across every industry and geography, we’ve seen that most transformation initiatives are designed to achieve one of four business outcomes:

1. Enriching employee experiences: Using AI to streamline or automate repetitive, mundane tasks can allow your employees to dive into more complex, creative, and ultimately more valuable work.
2. Reinventing customer engagement: AI can create more personalized, tailored customer experiences, delighting your target audiences while lightening the load for employees.
3. Reshaping business processes: Virtually any business process can be reimagined with AI, from marketing to supply chain operations to finance, and AI is even allowing organizations to go beyond process optimization and discover exciting new growth opportunities.
4. Bending the curve on innovation: AI is revolutionizing innovation by speeding up creative processes and product development, reducing the time to market and allowing companies to differentiate in an often crowded field.

In this blog, we’ve collected more than 200 of our favorite real-life examples of how organizations are embracing Microsoft’s proven AI capabilities to drive impact and shape today’s platform shift to AI. We hope you find an example or two that can inspire your own transformation journey.

IDC InfoBrief: sponsored by Microsoft, 2024 Business Opportunity of AI, IDC# US52699124, November 2024 

Enriching employee experiences

Generative AI is truly transforming employee productivity and well-being. Our customers tell us that by automating repetitive, mundane tasks, employees are freed up to dive into more complex and creative work. This shift not only makes the work environment more stimulating but also boosts job satisfaction. It sparks innovation, provides actionable insights for better decision-making, and supports personalized training and development opportunities, all contributing to a better work-life balance. Customers around the world have reported significant improvements in employee productivity with these AI solutions:

1. Access Holdings Plc adopted Microsoft 365 Copilot, integrating generative AI into daily tools and, as a result, writing code now takes two hours instead of eight, chatbots launch in 10 days instead of three months, and presentations are prepared in 45 minutes instead of six hours.
2. Adobe is connecting Adobe Experience Cloud workflows and insights with Microsoft 365 Copilot to deliver generative AI-powered capabilities that enable marketers to increase collaboration, efficiency, and creativity.
3. Amadeus empowers its teams to focus their time and skills on value-added tasks with Microsoft 365 Copilot by summarizing email threads, chat, or transcripts, and summing up information from diverse sources.
4. ANZ has invested in Microsoft 365 Copilot, GitHub Copilot, and Copilot in Microsoft Edge to boost productivity and innovation across its workforce. 
5. Asahi Europe & International (AEI) has adopted Microsoft 365 Copilot, saving employees potentially 15% of time previously spent on administrative tasks.
6. AXA developed AXA Secure GPT, a platform powered by Azure OpenAI Service that empowers employees to leverage the power of generative AI while targeting the highest level of data safety and responsible use of the tool.
7. Axon Enterprise developed a new AI tool with Azure OpenAI Service called Draft One, resulting in an 82% decrease in time spent on reports, which freed up officers to engage more with their community.
8. Aztec Group enhanced productivity and client experience by trialing Microsoft 365 Copilot with 300 staff, uncovering “unlimited” use cases and plans for a wider rollout.
9. Bader Sultan & Bros. Co. W.L.L implemented Microsoft 365 Copilot to enhance employee productivity and speed up customer response times.
10. Bancolombia is using GitHub Copilot to empower its technical team, achieving a 30% increase in code generation, boosting automated application changes to an average of 18,000 per year, with a rate of 42 productive daily deployments. 
11. BaptistCare Community Services is using Microsoft 365 Copilot to save employees time as they navigate workforce shortage challenges allowing them to focus more on the people they care for.
12. Barnsley Council was recognized as “Double Council of the Year in 2023” for its implementation of Microsoft 365 Copilot, which modernized operations and reduced administrative tasks, leading to improved job satisfaction and increased creativity.
13. BlackRock purchased more than 24,000 Microsoft 365 Copilot licenses spanning all employees, functions, and locations, helping improve the Copilot experience, including co-developing new features and functions.
14. British Heart Foundation is testing Microsoft 365 Copilot and in its initial test, users estimate that Microsoft 365 Copilot could save them up to 30 minutes per day.
15. Buckinghamshire Council deployed Microsoft 365 Copilot with staff reporting productivity improvements, quality enhancements, and time savings which are enabling the different teams to do more with less. 
16. Campari Group adopted Microsoft 365 Copilot to help employees integrate it into their workflow, resulting in time savings of about two hours a week from the support of routine activities such as email management, meeting preparation, content creation, and skill acquisition.
17. Capita is using GitHub Copilot for productivity improvements as well as improvements in developer satisfaction, recruitment, and retention.
18. CDW used Microsoft 365 Copilot to improve work quality for 88% of users, enabling 77% to complete tasks faster, and increasing productivity for 85% of users.
19. Chi Mei Medical Center is lightening workloads for doctors, nurses, and pharmacists with a generative AI assistant built on Azure OpenAI Service. 
20. E.ON is focused on Germany’s energy transition, leveraging Microsoft 365 Copilot to manage the complex grid in real-time, increasing productivity and efficiency for its workforce.
21. Enerijisa Üretim has adopted Microsoft 365 Copilot to streamline meeting summaries, reformat documents, and compile reports, enabling employees to concentrate on more strategic and fulfilling activities instead of spending six hours in meetings.
22. EPAM is deploying Microsoft 365 Copilot to consolidate information and generate content and documents. 
23. Farm Credit Canada implemented Microsoft 365 Copilot which resulted in time savings on routine tasks for 78% of users, with 30% saving 30 to 60 minutes per week and 35% saving over an hour per week, allowing employees to focus on more value-added tasks.
24. Finastra used Microsoft 365 Copilot to automate tasks, enhance content creation, improve analytics, and personalize customer interactions, with employees citing a 20% to 50% time savings.
25. Four Agency Worldwide increased employee productivity using Microsoft 365 Copilot to generate ideas for creative work and support administrative-heavy processes, data analysis, and report generation, allowing staff to focus on outreach and less time doing paperwork.
26. Goodwill of Orange County developed an AI-powered app using Azure AI capabilities to help more people, including those with developmental, intellectual, and physical disabilities, work in unfilled e-commerce positions.
27. Honeywell employees are saving 92 minutes per week—that’s 74 hours a year! Disclaimer: Statistics are from an internal Honeywell survey of 5,000 employees where 611 employees responded.
28. Insight employees using Copilot are seeing four hours of productivity gained per week from data summarization and content creation.
29. Joos uses Microsoft 365 Copilot to grow its brand with worldwide collaboration by streamlining meetings, optimizing presentations, and improving communications.
30. Kantar is harnessing the power of Microsoft 365 Copilot by reducing costly, time-consuming IT processes and boosting productivity for employees.
31. KPMG Australia is using Microsoft Azure OpenAI Service, Azure AI Search, and Microsoft 365 Copilot to perform advanced text analysis of dozens of client source documents to identify full or partial compliance, or noncompliance, in a fraction of the time required for manual assessments.
32. LGT is launching Microsoft Copilot LGT to improve efficiency, showing users save an average of an hour a week even in the pilot phase. 
33. Lotte Hotels & Resorts has been creating a new work culture that allows employees to work more efficiently and focus on the nature of the work by adopting Microsoft Power Platform for automation.
34. MAIRE is leveraging Microsoft 365 Copilot to automate routine tasks, saving more than 800 working hours per month, freeing up engineers and professionals for strategic activities while supporting MAIRE’s green energy transition by reducing their carbon footprint.
35. McDonald’s China chose Microsoft Azure AI, GitHub Copilot, and Azure AI Search to transform its operations, resulting in a significant increase in AI adoption, consumption, and retention from 2,000 to 30,000 employee transactions monthly.
36. McKnight Foundation adopted Microsoft 365 Copilot for all staff, saving time, increasing productivity, and freeing space to focus on strategic priorities.
37. Morula Health is using Microsoft 365 Copilot to enhance productivity, streamline medical writing tasks, and ensure data security, ultimately improving efficiency and client satisfaction. 
38. Motor Oil Group is achieving remarkable efficiency gains by integrating Microsoft 365 Copilot into its workflows, with staff spending minutes on tasks that used to take weeks. 
39. Nagel-Group uses Azure OpenAI Service to help employees quickly access information which saves time, creates efficiency and transparency, and leads to higher-quality answers overall.
40. National Australia Bank is leveraging Microsoft 365 Copilot for daily productivity and data analysis and insights and Microsoft Security Copilot to quickly analyze millions of security event logs and allow engineers to focus on more important areas.
41. NFL Players Association integrated Azure AI Services and Azure App Service into their video review process, reducing review time by up to 73%, significantly increasing efficiency and enhancing player safety through consistent rule enforcement.
42. O2 Czech Republic boosts productivity and streamlines meetings with Microsoft 365 Copilot, revolutionizing how information is shared and making automation a part of daily work.
43. Onepoint developed a secure conversational agent based on Azure OpenAI Service, which delivers productivity gains of between 10% and 15% across all business lines.
44. Orange Group has more than 40 use cases with Azure OpenAI Service and GitHub Copilot across business functions to support employees in their day-to-day tasks, enabling them to concentrate on higher value-added activities.
45. Oxford University Hospitals NHS Foundation Trust implemented Microsoft 365 Copilot to improve staff report productivity by saving one to two hours a week, or simple formatting tasks down to a matter of seconds, enabling more resources to deliver frontline services.
46. PA Consulting transformed its sales operations with Microsoft 365 Copilot, so its people can invest more time on the activities that have the biggest impact for clients and maximize the strategic value they provide. 
47. Petrobras used Azure OpenAI Service to create ChatPetrobras, which is streamlining workflows, reducing manual tasks, and summarizing reports for its 110,000 employees.
48. Petrochemical Industries Company automates work processes to save time with Microsoft 365 Copilot from weeks to days, hours to seconds.
49. PKSHA Technology is optimizing their time on critical work by increasing efficiency in meeting preparations, data analytics, and ideation with the help of Microsoft 365 Copilot.
50. Providence has collaborated with Nuance and Microsoft to accelerate development and adoption of generative AI-powered applications, helping improve care quality and access, and reduce physician’s administrative workloads. 
51. RTI International adopted Microsoft 365 Copilot to gain productivity wherever possible, allowing staff to focus on their areas of expertise, delivering even better science-backed solutions for clients.
52. Sandvik Coromant is using Microsoft 365 Copilot for Sales to drive efficiency and accuracy, shaving at least one minute off each transaction, allowing sellers and account managers to focus their expertise on responding to customers’ needs with analysis, creativity, and adaptability.
53. Sasfin Bank built a solution on Microsoft Azure that centralized 20,000 documents to analyze contract clauses and provide real-time snapshots, moving guesswork into data-driven decision-making.
54. Scottish Water implemented Microsoft 365 Copilot reducing mundane tasks to a minimum, and thus freeing up time for employees to work on the more meaningful tasks.
55. Shriners Children’s developed an AI platform allowing clinicians to easily and securely navigate patient data in a singular location, enhancing patient care, and improving the efficiency of their healthcare services. 
56. Siemens is leveraging Azure OpenAI Service to improve efficiency, cut downtime, and address labor shortages.
57. Softchoice employees are experiencing firsthand how Microsoft 365 Copilot can transform daily workflows, realizing productivity gains of 97% reduction in time spent summarizing technical meetings and up to 70% less time spent on content creation.
58. Syensqo utilized Azure OpenAI Service to develop a custom AI chatbot in three months, which improved their internal data management, decision-making, and overall efficiency.
59. Teladoc Health uses Microsoft 365 Copilot to revolutionize its telehealth operations, automating routine tasks, boosting efficiency, and increasing productivity.
60. Telstra developed two cutting-edge generative AI tools based on Azure OpenAI Service: 90% of employees are using the One Sentence Summary tool which resulted in 20% less follow-up customer contact and 84% of customer service agents using the Ask Telstra solution.
61. Topsoe achieved 85% AI adoption among office employees in seven months, significantly enhancing productivity and business processes.
62. Torfaen County Borough Council utilized Microsoft 365 Copilot to streamline back-office processes, resulting in significant time savings and enhanced productivity for both business and children’s services teams, with further rollouts planned.
63. Trace3 leveraged Microsoft 365 Copilot to streamline and enhance processes across the business and with clients, such as reducing the time it takes human resources (HR) recruiting managers to respond to applicants within a couple of days instead of several weeks.
64. Unilever is reinventing their marketing process with Copilot, saving time on briefing tasks, automatically pulling in relevant market data, content and insights to accelerate campaign launches. 
65. Uniper SE implemented Microsoft 365 Copilot to reduce time spent on manual and repetitive tasks, and help workers focus on more pressing work, such as developing enhanced solutions to speed up the energy transition.
66. Unum Group built a custom AI application to search 1.3 terabytes of data with 95% accuracy using Azure OpenAI Service. 
67. Virgin Atlantic adopted Microsoft 365 Copilot and GitHub Copilot and is seeing real business benefits, including productivity improvements, enabling new ways of working.
68. Visier built a generative AI assistant that leverages Azure AI and Azure OpenAI Services to deliver workforce analytics and actionable insights for more than 50,000 customers.
69. Virtual Dental Care developed an AI application Smart Scan that leverages Azure to reduce paperwork for mobile dental clinics in schools by 75% and frees dentists to devote more time to patient care.
70. Zakladni Skola As Hlavkova adopted Microsoft 365 Copilot and saw a 60% improvement in handling administrative documents, decreased lesson preparation from hours to few minutes, increased inclusivity, and enhanced communication with students and parents.

Reinventing customer engagement

We’ve seen great examples of how generative AI can automate content creation, ensuring there’s fresh and engaging materials ready to go. It personalizes customer experiences by crunching the numbers and boosting conversion rates. It makes operations smoother, helping teams launch campaigns faster. Plus, it drives innovation, crafting experiences that delight customers while lightening the load for staff. Embracing generative AI is key for organizations wanting to reinvent customer engagements, stay ahead of the game, and drive both innovation and efficiency.

71. Absa has adopted Microsoft Copilot to streamline various business processes, saving several hours on administrative tasks each day.
72. Adobe leverages Azure to streamline the customer experience, harnessing the power of the connected cloud services and creating a synergy that drives AI transformation across industries.
73. Acentra Health developed Medscribe, a web application that uses Azure OpenAI Service to generate draft letters in a secure, HIPAA-compliant enclave that responds to customer appeals for healthcare services within 24 hours, reducing the time spent on each appeal letter by 50%.
74. Alaska Airlines is using Azure, Microsoft Defender, and GitHub to ensure its passengers have a seamless journey from ticket purchase to baggage pickup and started leveraging Azure OpenAI Service to unlock more business value for its customer care and contact centers.
75. Ally Financial is using Azure OpenAI Service to reduce manual tasks for its customer service associates, freeing up time for them to engage with customers. 
76. BMW Group optimizes the customer experience connecting 13 million active users to their vehicles with the MyBMW app on Azure, which supports 450 million daily requests and 3.2 terabyte (TB) data processing.
77. Boyner has tripled its e-commerce performance using Azure, seeing a rise in customer satisfaction, engagement, conversion rate, and revenue.
78. Bradesco Bank integrated Azure to its virtual assistant, BIA, resulting in reduced response time from days to hours, improving operational efficiency and client satisfaction.
79. Capgemini Mexico integrated GitHub Copilot to support scalable AI implementations which has led to improved customer experiences and increased efficiency.
80. Capitec Bank uses Azure OpenAI Service and Microsoft 365 Copilot, enabling their AI-powered chatbot to assist customer service consultants in accessing product information more efficiently, saving significant time for employees each week.
81. Cdiscount is leveraging GitHub Copilot and Azure OpenAI Service to enhance developer efficiency, optimize product sheet categorization, and improve customer satisfaction.
82. Cemex used Azure OpenAI Service to launch Technical Xpert, an AI tool used by sales agents to provide instant access to comprehensive product and customer solution information, significantly reducing search time by 80%. 
83. Chanel elevated their client experience and improved employee efficiency by leveraging Microsoft Fabric and Azure OpenAI Service for real-time translations and quality monitoring.
84. City of Burlington created two AI-powered solutions: MyFiles system using Microsoft Power Platform for building permits, and CoBy, an around-the-clock customer support assistant using Microsoft Copilot Studio.
85. City of Madrid created an AI virtual assistant with Azure OpenAI Service offering tourists accurate, real-time information and personalized responses in more than 95 languages.
86. Cognizant is making performance management more effective and meaningful with Azure Machine Learning to help clients across industries envision, build, and run innovative digital enterprises.
87. Coles Group has leveraged Azure to enhance its digital presence and improve customer engagement, rolling out new applications to its stores six times faster without disrupting workloads.
88. Commercial Bank of Dubai used Azure to upgrade its application infrastructure, improving transaction security and speed so individual customers can now open an account and start banking in about two minutes.
89. Dubai Electricity and Water Authority has significantly improved productivity and customer satisfaction by integrating multiple Microsoft AI solutions, reducing task completion time from days to hours and achieving a 98% customer happiness rate.
90. Elcome uses Microsoft 365 Copilot to improve the customer experience, reducing response times from 24 hours to eight hours.
91. elunic developed shopfloor.GPT based on Azure OpenAI Service, leading to increased productivity for customers, saving 15 minutes per request.
92. Estée Lauder Companies is leveraging Azure OpenAI Service to create closer consumer connections and increase speed to market with local relevancy.
93. First National Bank (FNB) is using Microsoft 365 Copilot for Sales to help bankers create professional, thoughtful emails in 13 native South African languages to enhance customer interactions, streamline communications, and reinforce its commitment to innovation and customer service. 
94. Flora Food Group migrated to Microsoft Fabric to offer more detailed and timely insights to its customers, enhancing service delivery and customer satisfaction.
95. Groupama deployed a virtual assistant using Azure OpenAI Service that delivers reliable, verified and verifiable information, and boasts an 80% success rate.
96. International University of Applied Sciences (IU) adopted Azure OpenAI Service to revolutionize learning with a personalized study assistant that can interact with each student just like a human would.
97. Investec is using Microsoft 365 Copilot for Sales to enhance the bank’s client relationships, estimating saving approximately 200 hours annually ultimately boosting sales productivity and delivering personalized, seamless customer experience. 
98. Linum is using Azure to train their text-to-video models faster and more efficiently without losing performance or wasting resources.
99. Lumen Technologies is redefining customer success and sales processes through the strategic use of Microsoft 365 Copilot, enhancing productivity, sales, and customer service in the global communications sector.
100. McKinsey & Company is creating an agent to reduce client onboarding process by reducing lead time by 90% and administrative work by 30%.
101. Meesho leveraged Azure OpenAI Service and GitHub Copilot to enhance customer service and software development, resulting in a 25% increase in customer satisfaction scores and 40% more traffic on customer service queries.
102. Milpark Education integrated Microsoft Copilot and Copilot Studio and in just four months, improved efficiency and accuracy of student support, decreasing the average resolution time by 50% and escalations by more than 30%.
103. NC Fusion chose a comprehensive Microsoft solution to make marketing engagement activities easier and accurately target the best audience segments.
104. Medgate, a telehealth subsidiary of Otto Group, developed a medical Copilot powered by Azure OpenAI Service that summarizes consultations, supports triage, and provides real-time translations. 
105. Pacific Gas & Electric built a chatbot using Microsoft Copilot Studio that saves $1.1 million annually on helpdesk support. 
106. Pockyt is using GitHub Copilot and anticipates a 500% increase in productivity in the medium to long term as they continue adapting AI and fine-tuning their software development life cycle.
107. South Australia Department for Education launched an AI-powered educational chatbot to help safeguard students from harmful content while introducing responsible AI to the classrooms.
108. Sync Labs is using Azure to create AI-powered solutions that have led to a remarkable 30x increase in revenue and a 100x expansion of their customer base.
109. Syndigo is using Azure to accelerate digital commerce for its customers by more than 40% and expand its customer base.
110. Telkomsel created a virtual assistant with Azure OpenAI Service, resulting in a leap in customer self-service interactions from 19% to 45%, and call volume dropped from 8,000 calls to 1,000 calls a day.
111. Torrens University chose to use Azure OpenAI to uplift its online learning experience, saving 20,000 hours and $2.4 million in time and resources.
112. Trusting Social integrated Azure services to launch AI-driven agents that are changing how banks function and transforming their customer’s banking experience.
113. University of California, Berkeley used Azure OpenAI Service to deploy a custom AI chatbot that supports student learning and help students with complex coursework.
114. University of Sydney created a self-serve AI platform powered by Azure OpenAI Service, to enable faculty to build custom chatbots for enhancing student onboarding, feedback, career simulation, and more.
115. Van Lanschot Kempen is using Microsoft 365 Copilot to reduce the time needed for daily tasks, freeing up time to invest in that crucial personal connection.
116. Virgin Money built an award-winning virtual assistant using Copilot Studio to help build customers’ confidence in their digital products and services.
117. VOCALLS automates over 50 million interactions per year, resulting in a 78% reduction in average handling time aside from a 120% increase in answered calls.
118. Vodafone Group is leveraging Microsoft’s AI solutions, including Azure AI Studio, OpenAI Service, Copilot, and AI Search, to achieve a 70% resolution rate for customer inquiries through digital channels and reduce call times by at least one minute.
119. Walmart is using Azure OpenAI Service to deliver a helpful and intuitive browsing experience for customers designed to serve up a curated list of the personalized items a shopper is looking for.
120. Weights & Biases created a platform which runs on Azure that allows developers to keep records, log successes and failures, and automate manual tasks.
121. World2Meet is providing better customer service and operations with a new virtual assistant powered by Azure.
122. Xavier College is modernizing its student information systems on Microsoft Dynamics 365 and Azure to unlock powerful insights, fostering innovation and data-driven decision making.
123. Zavarovalnica Triglav implemented Dynamics 365 and Azure OpenAI Service to streamline its operations with automated responses and smart rerouting of customer enquiries.

Reshaping business processes

Transforming operations is another way generative AI is encouraging innovation and improving efficiency across various business functions. In marketing, it can create personalized content to truly engage different audiences. For supply chain management, it can predict market trends so companies can optimize their inventory levels. Human resources departments can speed up the hiring process, while financial services can use it for fraud detection and risk assessments. With generative AI, companies are not just refining their current processes, they’re also discovering exciting new growth opportunities.

124. Accelleron used Microsoft Power Platform to support numerous business applications and simplify processes for service agents and employees, resulting in the onboard of new agents in 30 minutes, compared to two days for other solutions.
125. Accenture developed an AI-powered financial advisor that leverages RISE with SAP on Azure to enhance their infrastructure and integrate financial data.
126. Atomicwork leverages Azure OpenAI to bring together three power capabilities: a conversational assistant, a modern service management system, and a workflow automation platform.
127. Blink Ops fully embraced generative AI to build the world’s first Security Automation Copilot with more than 8,000 automated workflows to help any Security/IT task through prompts.
128. Chalhoub Group is using Microsoft Fabric to modernize its data analytics and streamline its data sources into one platform, increasing agility, enhancing analytics, and accelerating processes.
129. Cineplex is developing innovative automation solutions for finance, guest services, and other departments, saving the company more than 30,000 hours a year in manual processing time.
130. ClearBank moved its services to Azure to gain scalability and efficiency, pushing out 183% more monthly system releases, gaining both scalability and efficiency.
131. Danske Statsbaner increases productivity up to 30% with help from Microsoft AI solutions.
132. Eastman implemented Microsoft Security Copilot realizing the benefits of accelerated upskilling, step-by-step guidance for response, and faster threat remediation.
133. Fast Shop migrated to Azure creating a self-service culture of access to data, eliminating delays, reducing costs, and increasing leadership satisfaction with data while providing more agility in reporting.
134. Florida Crystals adopted a value-added solution across Microsoft products including Microsoft 365 Copilot to reduce telecom expenses and automate industrial process controls.
135. GHD is reinventing the request for proposal (RFP) process in construction and engineering with Microsoft 365 Copilot.
136. GovDash is a software as a service (SaaS) platform that leverages AI to streamline the entire business development lifecycle for government contracting companies using Azure OpenAI Service.
137. Grupo Bimbo is deploying Microsoft’s industrial AI technologies to modernize its manufacturing processes, optimizing production and reducing downtime, driving significant cost savings, and empowering global innovation.
138. Insight Canada implemented Microsoft 365 Copilot to streamline business operations, realizing a 93% productivity gains in functions including sales, finance, and human resources.
139. Intesa Sanpaolo Group enhanced its cybersecurity with AI-enabled Microsoft Sentinel and Microsoft Security Copilot, resulting in faster threat detection, increased productivity, and reduced storage costs.
140. Kaya deployed a custom implementation of Dynamics 365 and Power BI to modernize its supply chain, leading to enhanced visibility, improved planning, and streamlined inter-department operations.
141. Lionbridge Technologies, LLC is using Azure and Azure OpenAI Service to accelerate its delivery times and improve quality, reducing project turnaround times by up to 30%.
142. LTIMindtree integrated Microsoft Security Copilot, offering automated incident response, integrated threat intelligence, and advanced threat analysis.
143. Mania de Churrasco used Azure, Microsoft Power Platform and Microsoft 365 to achieve high efficiency, security, and scalability in its operations, in addition to improving its data intelligence, which indirectly participated in a 20% increase in sales year on year.
144. National Bank of Greece built an Azure-powered Document AI solution to transform its document processing, improving the bank’s accuracy to 90%. 
145. Nest Bank has revolutionized its operations by integrating Microsoft 365 Copilot and Azure OpenAI Service, resulting in doubled sales and increased daily transactions from 60,000 to 80,000 showcasing the transformative impact of generative AI in the financial sector.
146. Network Rail modernized their data analytics solution with Azure, helping engineers understand data 50% faster than before and improve efficiency, passenger experiences, and safety—all while saving costs.
147. Nsure developed an AI-powered agent that uses Copilot Studio and Power Automate to reduce manual processing time by 60% while also reducing associated costs by 50%.
148. Oncoclínicas implemented Azure to transform its entire data ecosystem with a web portal and mobile application that performs all image processing and storage.
149. Pacífico Seguros has adopted Microsoft Security Copilot to optimize its security operations and anticipate and neutralize threats more efficiently and effectively.
150. Parexel adopted Azure Databricks and Power BI, achieving an 85% reduction in data engineering tooling costs, a 30% increase in staff efficiency, and a 70% reduction in time to market for data product delivery.
151. Paysafe used Microsoft 365 Copilot to streamline meetings, information management, and document creation, addressing language barriers, eliminating time-consuming tasks, and boosting creativity along the way.
152. Planted is integrating Azure OpenAI Service to manage everyday tasks more efficiently and facilitate the search for information for innovative process development.
153. Presidio realized dramatic productivity gains saving 1,200 hours per month on average for the employees using Microsoft 365 Copilot and created 70 new business opportunities.
154. Qatar Charity used Copilot Studio to increase its call center efficiency, reducing average handle time by 30%, increased customer satisfaction by 25%, and achieved a 40% reduction in IT maintenance costs.
155. Saphyre uses Azure and AI to provide an intelligent cloud-based solution that automates and streamlines financial trading workflows around client and counterparty life cycle management, reducing manual efforts by 75%. 
156. Swiss International Air Lines migrated and modernized with Azure, achieving up to 30% cost savings, a remarkable boost in platform stability along with enhanced security visibility.
157. ZEISS Group uses Microsoft Fabric to create a secure and trusted data supply chain that can be shared effortlessly across a range of business units.
158. ZF Group builds manufacturing efficiency with more than 25,000 apps and 37,000 unique active users on Microsoft Power Platform.

Bending the curve on innovation

Generative AI is revolutionizing innovation by speeding up creative processes and product development. It’s helping companies come up with new ideas, design prototypes, and iterate quickly, cutting down the time it takes to get to market. In the automotive industry, it’s designing more efficient vehicles, while in pharmaceuticals, it’s crafting new drug molecules, slashing years off research and development (R&D) times. In education, it transforms how students learn and achieve their goals. Here are more examples of how companies are embracing generative AI to shape the future of innovation.

159. Air India has incorporated Microsoft 365 Copilot into multiple departments, unlocking a new realm of operational insights that not only provides critical data on flight punctuality and operational hurdles, but also empowers proactive, collaborative decision making.
160. Agnostic Intelligence deployed Azure OpenAI Service to eliminate time-consuming tasks, saving users up to 80% of their time, and enabling IT managers to focus on innovation and quality assurance.
161. Albert Heijn is using Azure OpenAI Service for everything from customer personalization to demand forecast and food waste projects, making it easier for its customers to change their lifestyle.
162. Amgen is using Microsoft 365 Copilot to boost productivity and has the potential to speed up drug development and support advancements in their business processes.
163. APEC leverages Azure and deep neural network algorithms to develop an app that enables healthcare providers to capture retinal images, increasing the accuracy to identify Retinopathy of Prematurity (RoP) to 90%.
164. ASOS is using Azure AI Studio to help customers discover new looks with genuine shopping insights, personalized conversations, naturalism, and even humor to enliven the shopping journey.
165. Auburn University is incorporating Microsoft Copilot to promote AI literacy, accessibility, and collaboration, with the aim to expand educational and economic opportunities for its entire academic community with AI-centric tools.
166. B3 launched an AI assistant using Azure OpenAI Service that aids 10,000 users a day to answer Brazilians’ questions about how to start investing.
167. Basecamp Research aims to build the world’s largest database of national biodiversity and apply AI and machine learning to advance bioscience.
168. Bayer is using Microsoft Copilot to contribute to feeding a growing global population and helping people lead healthier, disease-free lives.
169. Brembo leveraged Azure OpenAI Service to develop ALCHEMIX, a solution to generate innovative compounds for its brake pads, drastically reducing the development time of new compounds from days to mere minutes.
170. Canary Speech can now train new vocal models in as little as two months and handle millions of transactions per month with Azure. 
171. CapitaLand simplified internal processes increasing efficiency to more than 10,000 man-days saved per year and deployed Azure OpenAI Service to build the first AI hospitality chatbot for its lodging business.
172. Cassidy is using Azure OpenAI Service to enhance efficiency across various industries, supporting more than 10,000 companies.
173. Coca-Cola is implementing Azure OpenAI Service to develop innovative generative AI use cases across various business functions, including testing how Microsoft 365 Copilot could help improve workplace productivity.
174. Denso is developing “human-like” robots using Azure OpenAI Service as the brain to help robots and humans work together through dialogue.
175. eFishery is using Azure OpenAI for farmers to get the data and insights on fish and shrimp farming, including more precise feeding and water quality monitoring.
176. EY developed an application that automatically matches and clears incoming payments in SAP, resulting in an increase from 30% to 80% in automatically cleared payments and 95% matched payments, with estimated annual time savings of 230,000 hours globally.
177. FIDO is using Azure OpenAI Service to develop an AI tool that uses sound to pinpoint leaky pipes, saving precious drinking water. 
178. Georgia Tech is using Azure OpenAI Service to enhance the electric vehicle (EV) charging infrastructure, achieving rapid data classification and predictive modeling, highlighting the reliability of networked chargers over non-networked ones.
179. GigXR developed a solution to create the intelligence for specific AI patients using Azure OpenAI Service and other Azure services.
180. GoTo Group is significantly enhancing productivity and code quality across its engineering teams by adopting GitHub Copilot, saving over seven hours per week and achieved a 30% code acceptance rate.
181. GovTech used Azure OpenAI Service to create LaunchPad, sparking more than 400 ideas and 20 prototypes, laying the foundation for the government to harness the power of generative AI.
182. H&R Block is using Azure AI Studio and Azure OpenAI Service to build a new solution that provides real-time, reliable tax filing assistance.
183. Haut.AI provides skin care companies and retailers with customizable, AI-based skin diagnostic tools developed with the help of Microsoft AI.
184. Helfie is building a solution that caters to healthcare providers who can arm their patients with an application to more quickly and accurately access the care they need.
185. Hitachi will implement Azure Open AI Service, Microsoft 365 Copilot, and GitHub Copilot to create innovative solutions for the energy, mobility, and other industries.
186. Icertis is providing AI-based tools that will recognize contract language and then build algorithms to automatically choose the right approach based on the content of the contract.
187. ITOCHU is using Azure OpenAI Service and Azure AI Studio to evolve its data analytics dashboard into a service that provides immediate recommendations by automatically creating evidence-based product proposals.
188. IU International University of Applied Sciences (IU) is using the power of Azure OpenAI Service to develop Syntea, an AI avatar integrated into Microsoft Teams and Microsoft 365 Copilot, making learning more personalized, autonomous, and flexible.
189. Khan Academy has partnered with Microsoft to bring time-saving and lesson-enhancing AI tools to millions of educators.
190. Lufthansa Group developed an animated 3D avatar called Digital Hangar to help guide passengers from initial travel inspiration to flight booking through an exchange with an avatar in natural language.
191. Mitsubishi Heavy Industries is using Azure OpenAI Service to help accelerate digital innovation in power plants.
192. Molslinjen has created an AI analytics toolbox that has reduced fuel emissions, improved customer satisfaction, and brought in millions of additional revenue.
193. Novo Nordisk recently published initial results with predictive AI models for advanced risk detection in cardiovascular diseases, including an algorithm that can predict patients’ cardiovascular risk better than the best clinical standards. 
194. Paige.AI​ is using AI and Azure to accelerate cancer diagnoses with data from millions of images.
195. Pets at Home created an agent to help its retail fraud detection team investigate suspicious transactions.  
196. Plan Heal is using Microsoft AI to create solutions that enable patients to monitor and report health metrics so care providers can better serve them.
197. Pacific Northwest National Laboratory (PNNL) is testing a new battery material that was found in a matter of weeks, not years, as part of a collaboration with Microsoft.
198. Rijksmuseum is harnessing the power of Copilot to make art accessible at scale by joining forces with Microsoft to improve and expand the art experience for blind and low-vision community members.
199. Royal National Institute of Blind People is using Azure AI Services to develop an AI-based solution that quickly and accurately converts letters to braille, audio, and large print formats.
200. Schneider Electric provides productivity-enhancing and energy efficiency solutions and is using a whole suite of AI tools to hasten its own innovation and that of its customers.
201. SPAR ICS created an award-winning, AI-enabled demand forecasting system achieving 90% inventory prediction accuracy.
202. Suzuki Motor Corporation is adopting Azure OpenAI Service for data security, driving company-wide use with five multipurpose apps.
203. Tecnológico de Monterrey created a generative AI-powered ecosystem built on Azure OpenAI Service with the goal to personalize education based on the students’ needs, improve the learning process, boost teachers’ creativity and save time on tedious tasks.
204. TomTom is using Azure OpenAI Service, Azure Cosmos DB, and Azure Kubernetes Service to revolutionize the driver experience.
205. Unilever is partnering with Microsoft to identify new digital capabilities to drive product innovation forward, from unlocking the secrets of our skin’s microbiome to reducing the carbon footprint of a multibillion-dollar business.
206. Unity used Azure OpenAI Service to build Muse Chat, an AI assistant that can guide creators through common questions and help troubleshoot issues to make game development easier.
207. University of South Florida is using Microsoft 365 Copilot to alleviate the burden of repetitive, time-consuming tasks so faculty and staff can spend this time creatively solving problems, conducting critical research, establishing stronger relationships with peers and students, and using their expertise to forge new, innovative paths. 
208. Visma has developed new code with GitHub Copilot, Azure DevOps, and Microsoft Visual Studio, as much as 50% faster, contributing to increased customer retention, faster time to market, and increased revenue.
209. Wallenius Wilhelmsen is implementing Microsoft 365 Copilot and using Microsoft Viva to drive sustainable adoption, streamlining processes, empowering better decision making, and cultivating a culture of innovation and inclusion.
210. Wipro is committed to delivering value to customers faster and improving the outcomes across the business by investing $1 billion in AI and training 200,000 employees on generative AI principles with Microsoft Copilot.

Next steps for AI transformation

Business Opportunity of AI

Get the study

• Check out the top 5 AI trends to watch from IDC and Microsoft.
• Explore Microsoft AI solutions.
• Assess your AI readiness.
• Build a strategic plan for AI.

The post How real-world businesses are transforming with AI appeared first on The Microsoft Cloud Blog.

To help guide organizations on their AI transformation journey, Microsoft recently commissioned a new study through IDC, The Business Opportunity of AI. IDC’s findings show that when organizations truly commit to and invest in AI, the return on investment (ROI) potential grows significantly.

According to IDC, the study’s findings reflect a tipping point as AI gains momentum across industries. As companies worldwide go deeper with AI, Microsoft customers continue to deploy innovative new solutions and discover how tools like Copilot can transform their day-to-day work. In telecommunications, Lumen Technologies estimates Copilot is saving sellers an average of four hours a week, equating to $50 million annually. In healthcare, Chi Mei Medical Center doctors now spend 15 minutes instead of an hour writing medical reports, and nurses can document patient information in under five minutes. Pharmacists are now able to double the number of patients they see per day. In retail, AI models help Coles predict the flow of 20,000 stock-keeping units to 850 stores with remarkable accuracy, generating 1.6 billion predictions daily.

IDC’s 2024 top 5 trends for AI

IDC’s findings align with what Microsoft is seeing as we work with companies across industries to deploy AI. We’ve highlighted more than 200 of our top AI customer stories to show a sampling of how AI is already driving impact today. Below is a look at the top trends we’re seeing in IDC’s study and the impact of those trends on organizations working with AI today.

#1 Enhanced productivity has become table stakes. Employee productivity is the No. 1 business outcome that companies are trying to achieve with AI. The study shows that 92% of AI users surveyed are using AI for productivity, and 43% say productivity use cases have provided the greatest ROI. While productivity is a top goal, generative AI use cases that are close behind include customer engagement, topline growth, cost management and product or service innovation — and nearly half of the companies surveyed expect AI to have a high degree of impact across all those areas over the next 24 months.

Customer snapshot:

At the global marketing and advertising agency dentsu, employees are already saving 15 to 30 minutes a day using Copilot for tasks such as summarizing chats, generating presentations and building executive summaries.

“Copilot has transformed the way we deliver creative concepts to our clients, enabling real-time collaboration. Agility, security and uniqueness are crucial, but our goal is to lead this transformation company-wide, from top to bottom.”

— Takuya Kodama, Business Strategy Manager at dentsu

#2 Companies are gravitating to more advanced AI solutions. In the next 24 months, more companies expect to build custom AI solutions tailored directly to industry needs and business processes, including custom copilots and AI agents. This shows a growing maturity in AI fluency as companies realize the value of out-of-the-box use cases and expand to more advanced scenarios.

Customer snapshot:

Siemens has developed the Siemens Industrial Copilot, which has eased the challenges caused by increasing complexity and labor shortages for dozens of customers in different industries.

“In full appreciation of GenAI’s transformational potential, it’s important to remember that production does not have an ‘undo’ button. It takes diligence and effort to mature AI to industrial-grade quality. The Siemens Industrial Copilot for Engineering significantly eases our customers’ workload and addresses the pressing challenges of skill shortages and increasing complexity in industrial automation. This AI-powered solution is a game-changer for our industry with over 50 customers already using it to boost efficiency and tackle labor shortages.”

— Boris Scharinger, AI Strategist at Siemens Digital Industries

#3 Generative AI adoption and value is growing across industries. Even though it is relatively new to the market, generative AI adoption is rapidly expanding — 75% of respondents report current usage up from 55% in 2023. The ROI of generative AI is highest in Financial Services, followed by Media & Telco, Mobility, Retail & Consumer Packaged Goods, Energy, Manufacturing, Healthcare and Education. Overall, generative AI is generating higher ROI across industries.

Customer snapshot:

Providence has leveraged AI to extend and enhance patient care, streamline processes and workflows and improve the effectiveness of caregivers.

“Whether we’re partnering with organizations on the leading edge of this technology — like Microsoft — and building bespoke solutions through Azure OpenAI Service, advancing clinical research to help cancer patients receive personalized and precise treatments faster, or ‘hitting the easy button’ and adopting established technologies like Microsoft 365 Copilot or DAX Copilot, we have successfully stayed on the forefront of this tech revolution. For example, physicians who use DAX Copilot save an average of 5.33 minutes per visit, and 80% of physicians have reported lower cognitive burden after using DAX Copilot.”

— Sarah Vaezy, EVP, Chief Strategy and Digital Officer at Providence

#4 AI leaders are seeing greater returns and accelerated innovation. While companies using generative AI are averaging $3.7x ROI, the top leaders using generative AI are realizing significantly higher returns, with an average ROI of $10.3. In addition to the enhanced business value, leaders are also on an accelerated path to build and implement new solutions — 29% of leaders implement AI in less than 3 months versus 6% of companies in the laggard category.

Customer snapshot:

Södra is an international forest industry group that processes forest products from 52,000 owners into renewable, climate-smart products for international market. Every day Södra collects and interprets climate impact data to make thousands of decisions for every part of the value chain.

“With innovative AI technology from Microsoft, our business experts and data scientists have been able to help make us more sustainable while also improving revenue significantly.”

— Cristian Brolin, Chief Digital Officer at Södra

#5 Looking ahead: Skilling remains a top challenge. Thirty percent of respondents indicated a lack of specialized AI skills in-house, and 26 percent say they lack employees with the skills needed to learn and work with AI. This dovetails with findings from the Microsoft and LinkedIn 2024 Work Trend Index Annual Report, which found that 55 percent of business leaders are concerned about having enough skilled talent to fill roles.

That is why over the past year we have helped train and certify over 14 million people in more than 200 countries in digital skills. And we are committed to working in partnership with governments, educational institutions, industry and civil society to help millions more learn to use AI.

Customer snapshot:

The University of South Florida (USF) is partnering with Microsoft to streamline processes and enhance innovation for all aspects of university operations with AI.

“We’re giving students a leg up to do amazing things with AI as part of tomorrow’s workforce. Our focus on generative AI not only drives operational efficiency but also empowers our community to unlock new levels of creativity and impact, further positioning USF as a leader in AI adoption, which includes being among the first universities in the nation to form a college dedicated to AI, cybersecurity and computing.”

— Sidney Fernandes, CIO & VP of Digital Experiences at University of South Florida

AI’s growing economic impact

While companies today are largely implementing out-of-the-box generative AI solutions and seeing significant ROI, more than half of those surveyed expect to build custom industry and line-of-business applications in the next 24 months — demonstrating that today’s ROI is quickly becoming tomorrow’s competitive edge.

“We are at an inflection point of autonomous agent development and are beginning an evolution from using just off-the-shelf assistants and copilots that support knowledge discovery and content generation to custom AI agents to execute complex, multistep workflows across a digital world,” says Ritu Jyoti, GVP/GM, AI and Data Research at IDC. “With responsible technology usage and workplace transformation, IDC predicts that business spending to adopt AI will have a cumulative global economic impact of $19.9 trillion through 2030 and drive 3.5% of global GDP in 2030.”

Key findings from IDC’s The Business Opportunity of AI study include:

• Generative AI usage jumped from 55% in 2023 to 75% in 2024.
• For every $1 a company invests in generative AI, the ROI is $3.7x.
• The top leaders using generative AI are realizing an ROI of $10.3.
• On average, AI deployments are taking less than 8 months and organizations are realizing value within 13 months.
• Within 24 months, most organizations plan to expand beyond pre-built AI solutions to advanced AI workloads that are customized or custom-built.
• The ROI of generative AI is highest in Financial Services, followed by Media & Telco, Mobility, Retail & Consumer Packaged Goods, Energy, Manufacturing, Healthcare and Education.
• 43% say productivity use cases have provided the greatest ROI.
• The primary way that organizations are monetizing AI today is through productivity use cases. In the next 24 months, a greater focus will be placed on functional and industry use cases.
• The top barrier when implementing AI is the lack of both technical and day-to-day AI skills.

Learn how to fuel your AI journey

IDC’s study, which included more than 4,000 business leaders and AI decision-makers around the world, also identifies the top barriers organizations face when implementing AI. As businesses integrate new solutions, they navigate important considerations such as data privacy, responsible use and the need for investment in both technology and skills.

No matter where you are in your cloud and AI transformation journey, Microsoft can help. To learn more about how customers across industries are shaping their AI transformation with Microsoft, please visit Microsoft’s AI in Action page. For more on how to get started in your AI transformation journey, visit Microsoft AI.

The post IDC’s 2024 AI opportunity study: Top five AI trends to watch appeared first on The Microsoft Cloud Blog.

An SLM focuses on specific AI tasks that are less resource-intensive, making them more accessible and cost-effective.¹ SLMs can respond to the same queries as LLMs, sometimes with deeper expertise for domain-specific tasks and at a much lower latency, but they can be less accurate with broad queries.² LLMs are an excellent choice for building your own enterprise custom agent or generative AI applications because of how capable they are.

Microsoft AI

Build the future of your business with AI

Explore solutions

Compare SLMs versus LLMs

Here are some criteria for each model type shown side-by-side to help you evaluate at a glance before diving deep into your due diligence and choosing one approach over another.

SLM and LLM functions

When comparing functions for small versus large language models, you should consider the balance between cost and performance. Smaller models typically require less computational power, reducing costs, but might not be well-suited for more complex tasks. Larger models offer superior accuracy and versatility but come with higher infrastructure and operational expenses. Evaluate your specific needs, like real-time processing, task complexity, and budget constraints, to make an informed choice.

You should also consider that SLMs can be fine-tuned to perform well in required tasks. Fine-tuning is a powerful tool to tailor advanced SLMs to your specific needs, using your own proprietary data. By fine-tuning an SLM, you can achieve a high level of accuracy for the particular use cases you require without needing to deploy an LLM that could be more expensive.  

For more complex tasks with a lot of edge cases, such as natural language queries or teaching a model to speak in a specific voice or tone, fine-tuning LLMs is a better solution. 

SLM and LLM features

Also be sure to consider features like computational efficiency, scalability, and accuracy. Smaller models often offer faster processing and lower costs, while larger models provide enhanced understanding and performance on complex tasks but require more resources. Evaluate your specific use cases and resource availability to help make an informed decision. 

SLM and LLM use cases

Carefully consider your specific use cases when comparing language models. Smaller models are ideal for tasks that require quick responses and lower computational costs, such as basic customer service chatbots or simple data extraction. On the other hand, large language models excel in more complex tasks requiring deep comprehension and nuanced responses, like advanced content generation or sophisticated data analysis. Aligning the model size with your specific business needs ensures you achieve both efficiency and effectiveness. 

SLM and LLM limitations

It’s also essential to consider limitations like computational requirements and scalability. Smaller models can be cost-effective and faster, but might not have the same nuanced understanding and depth of larger models. Larger models require significant computational resources, which can lead to higher costs and longer processing times. Balance these limitations against your specific use cases and available resources. 

This article touches on at-a-glance comparative information demonstrating the power and benefits of both SLMs and LLMs. With AI innovation accelerating at an intense pace involving different languages and scenarios, this rapid development will be sure to push the limits of both types of models—resulting in better, cheaper, and faster versions of current AI systems. This is particularly true for startups with limited resources where SLMs like Phi-3 open models will likely be the preferred, practical choice to leverage AI for their use cases.

Explore more resources on SLMs and LLMs

• Watch our AI in a Minute video about LLMs
• Explore our training: Introduction to large language models
• Read about the benefits of using SLMs in certain scenarios

AI learning hub

Get skilled up to power AI transformation

Start your journey

Our commitment to Trustworthy AI

Organizations across industries are leveraging Azure AI and Microsoft Copilot capabilities to drive growth, increase productivity, and create value-added experiences. 

We’re committed to helping organizations use and build AI that is trustworthy, meaning it is secure, private, and safe. We bring best practices and learnings from decades of researching and building AI products at scale to provide industry-leading commitments and capabilities that span our three pillars of security, privacy, and safety. Trustworthy AI is only possible when you combine our commitments, such as our Secure Future Initiative and our responsible AI principles, with our product capabilities to unlock AI transformation with confidence.      

Get started with Azure OpenAI Service

• See the latest Azure OpenAI Service news.
• Read more in our Azure AI services documentation.
• Read the latest AI and machine learning blogs.
• Listen to the podcast on Phi-3 with lead Microsoft researcher Sebastien Bubeck.

Learn more about AI solutions from Microsoft

• Explore Microsoft AI solutions to fuel your AI transformation.
• Learn how to build and optimize your strategic plan for AI with the AI Strategy Roadmap.
• Explore how customers are putting Microsoft AI to work.

¹Small Language Models (SLMs): The Next Frontier For The Enterprise, Forbes.

²Small Language Models vs. Large Language Models: How to Balance Performance and Cost-effectiveness, instinctools.

³Big is Not Always Better: Why Small Language Models Might Be the Right Fit, Intel.

The post Explore AI models: Key differences between small language models and large language models appeared first on The Microsoft Cloud Blog.

Germany stands at the forefront of a new era in AI, where strategic investments, public-private partnerships, and a focus on sustainable innovation converge to shape the future of industry and society. With significant governmental investment and growing adoption across key sectors, AI is poised to become a cornerstone of Germany’s economic transformation, ensuring global competitiveness and addressing critical labor shortages. 

The country’s rapidly advancing AI capabilities have been supported by significant investments and partnerships, with Microsoft playing a pivotal role. Earlier this year, Microsoft announced a €3.3 billion investment in Germany to expand its AI and cloud infrastructure over the next two years.¹ This initiative aims to double the AI infrastructure in the country and includes efforts to establish new datacenters in regions like North Rhine-Westphalia and Frankfurt, supporting key industries such as pharmaceuticals and energy through low-latency services. Microsoft is also focusing on sustainable operations, planning to power these facilities with renewable energy sources by 2025. 

In alignment with Germany’s broader national AI strategy, Microsoft has emphasized the importance of equipping workers with AI expertise, aiming to train up to 1.2 million people in new digital and AI capabilities by 2025.² This effort addresses labor gaps in sectors such as healthcare and manufacturing, which are adopting AI to increase efficiency and reduce operational costs. 

The recent Microsoft AI Tour event in Berlin, Germany, featured several innovative Germany organizations leveraging AI to advance to the cutting edge of their industries. Spanning from higher education to travel and manufacturing, these organizations present inspiring case studies and represent the tip of the iceberg in terms of what is possible at the intersection of German ingenuity and Microsoft AI technology.  

AI Office Hours

Build your generative AI fluency

Get started

IU International University of Applied Sciences (IU) prepares students for the future of work with AI-powered education  

IU International University of Applied Sciences (IU), Germany’s largest and fastest-growing university with more than 130,000 students, is using AI to advance its mission of democratizing higher education globally through a hybrid model of on-campus and online studies. By using the power of Microsoft Azure OpenAI Service, the university developed Syntea, an AI avatar integrated into Microsoft Teams and Microsoft 365 Copilot, making learning more personalized, autonomous, and flexible.   

Syntea’s impact has been profound, driving a 27% reduction in the time students need to complete online courses. In addition, IU is testing a new model within Syntea’s exam trainer function to improve grading while upholding its same high-quality standards and ensuring the solutions to be dependable. These innovations enhance the quality of education while making it more accessible and equitable. 

To further equip students for an AI-powered future, IU partnered with Microsoft to create the IU Copilot School featuring Microsoft. With the IU Copilot School, every student gains access to Microsoft 365 Copilot and in the future seamlessly to Syntea, fostering familiarity with AI tools that will shape tomorrow’s workforce. This initiative ensures AI is embedded across all study programs, preparing graduates to excel as knowledge workers in the evolving job market.   

Looking ahead, IU’s team of developers is exploring ways to extend the way of exam grading by leveraging the next generation of Syntea with advanced AI agents. IU’s Syntea Newskilling project leverages the power of AI-driven mentorship to redefine workplace learning and development by seamlessly integrating personalized upskilling and onboarding journeys directly within Microsoft Teams, empowering employees to achieve both individual and organizational success.

By seamlessly integrating AI into education and workforce preparation, International University of Applied Sciences (IU) is setting a bold precedent for how technology can redefine learning and unlock new opportunities for students and professionals alike. 

thyssenkrupp Automation Engineering alleviates labor shortage with Siemens Industrial Copilot 

Siemens and Microsoft have elevated the Siemens Industrial Copilot, now leveraging Azure OpenAI Service to meet rigorous manufacturing and automation demands at scale. The Industrial Copilot combines Siemens’ specialized industry expertise with advanced generative AI to address complex requirements, bringing new levels of efficiency and precision to industrial processes. 

thyssenkrupp Automation Engineering, a co-creation partner and one of more than 100 companies using the Siemens Industrial Copilot, plans to expand its use globally. The Industrial Copilot plays a critical role in thyssenkrupp’s response to the skilled labor shortage affecting Europe and the United States, enabling smoother operations even with less-experienced staff. 

A standout application is in thyssenkrupp’s battery quality systems, where the Industrial Copilot supports automated processes for battery assembly lines—key to the sustainable energy transition. For electric vehicle batteries, for example, the Industrial Copilot integrates sensors, cameras, and measurement systems to monitor quality at each production stage, detecting any issues that could affect reliability. By automating data management, sensor configuration, and stringent reporting requirements, the Industrial Copilot ensures precision in quality control, freeing engineers to focus on higher-value tasks. Additionally, the Industrial Copilot’s real-time problem-solving and documentation capabilities reduce downtime, while making machinery easier for early-career and unskilled workers to operate effectively.

Recognizing that technology oftentimes outpaces the machinery market, thyssenkrupp Automation Engineering is now the first to plan to use the Industrial Copilot globally, not only addressing labor challenges and advanced automation demands company-wide, but enabling their machines to keep pace with change into the future. 

Read more

• How AI is helping Siemens and thyssenkrupp bridge skilling gaps in manufacturing
• Siemens and Microsoft scale industrial AI

Ottobock revolutionizes prosthetics with AI for greater quality of life 

In pursuit of their mission to empower people with disabilities to move freely and independently, Ottobock is transforming individualized prosthetics and orthotics with the power of AI. For over a century, the company has faced a persistent challenge within the prosthetics and orthotics industry: achieving a perfect fit of body interfaces such as sockets or braces.   

With the help of the Microsoft Azure platform, Ottobock now analyzes detailed images of body parts to create customized body interfaces. This approach not only improves the precision and comfort of their devices, making high-quality solutions accessible to a broader range of patients, but also accelerates production while reducing costs. The streamlined process has further enabled less experienced technicians to deliver consistent, standardized products—helping alleviate the industry-wide shortage of skilled orthopedic technicians.   

Building on these successes, Ottobock is actively exploring additional AI applications to enhance the patient experience. With the potential of self-learning prosthetics on the horizon, Ottobock continues to elevate the quality of life for users by combining cutting-edge technology with compassionate care.

Lufthansa elevates customer experience and streamlines operations with AI-powered solutions  

Lufthansa Group continuously explores new ways to enhance premium travel experience of its passengers. This includes exploring possibilities of using generative AI. For example, in the Lufthansa Group Digital Hangar, an animated 3D avatar was developed as part of a test, guiding passengers from initial travel inspiration to flight booking through an exchange with an Avatar in natural language. Additionally, with the help of Microsoft’s generative AI, other applications were implemented, such as expanding the chat assistant, processing compensation claims, and automating website content creation.

Otto Group scales AI innovation, transforming e-commerce and healthcare 

Otto Group and its subsidiaries are harnessing the power of AI to accelerate innovation across industries. Group company OTTO has rolled out GitHub Copilot and the entire GitHub Platform for all software developers. This has enabled the teams to work numerous AI use cases simultaneously, which are used in e-commerce, live shopping, and by various subsidiaries of the Otto Group. One standout success is at Medgate, Otto Group’s leading telehealth subsidiary, which leverages AI-powered Copilot technology to address Germany’s healthcare challenges.   

Medgate developed a medical Copilot that transforms physician workflows and elevates patient care. Leveraging a custom-trained Azure speech and Azure OpenAI model, the Copilot summarizes consultations, supports triage, and provides real-time translations. In a medical chat powered by Azure OpenAI Service, physicians can enter keywords to generate follow-up questions and treatment recommendations. With this solution, Medgate reduces physicians’ administrative workload, giving doctors more time to focus on patient care—a critical advantage amid Germany’s shortage of 50,000 doctors and rising healthcare costs, which account for 12.8% of the nation’s GDP.   

The impact is measurable: case documentation times are cut by 10 to 20%, and AI-generated prompts reduce message drafting time by up to 40%. More than 60% of these AI-generated responses are utilized by doctors, demonstrating high acceptance and trust.

Otto Group’s successful integration of AI extends beyond Medgate, with the GitHub Cloud Platform enabling secure, scalable AI deployments across the Otto Group’s diverse operations. As Otto Group continues to lead in AI innovation, it exemplifies how technology can enhance both business performance and essential public services, driving meaningful impact across industries.

AI for everyone in Germany

As Germany cements its position as a global leader in AI, the ingenuity of its organizations continues to shape industries through transformative AI applications. From advancing education with AI-powered tools at IU International University of Applied Sciences to modernizing manufacturing at thyssenkrupp and enhancing healthcare services through Otto Group’s Medgate platform, these enterprises exemplify how AI can drive progress while addressing societal challenges.  

Microsoft’s investment in AI infrastructure and skilling reflects a long-term commitment to strengthening Germany’s digital capabilities and a shared goal to empower businesses and workers while fostering innovation that aligns with environmental sustainability and data security standards. Together, they are driving a future where AI not only serves as a tool for competitive advantage but also as a force for good—ensuring that Germany’s AI ecosystem becomes a model of innovation, resilience, and responsible development.

Find the resources to support your AI journey 

• Explore Microsoft’s AI solutions.
• Catch up on core concepts with AI Office Hours.
• Read the guide: An Introduction to Generative AI and Safety.

¹ The Federal Government, Investing in the digital future: Microsoft invests billions in Germany as an AI hub, February 15, 2024.

² Deutsche Welle, Germany: Microsoft to invest €3.3 billion in AI capacities, February 15, 2024.

The post German ingenuity meets the power of AI to shape the future of industries appeared first on The Microsoft Cloud Blog.

If you’ve read The AI Strategy Roadmap: Navigating the stages of value creation, you’re already familiar with the five drivers of AI value. This research paper and blog series explore what organizations need to succeed with AI, including establishing a clear strategy and securing senior leadership support. Our research found that AI success isn’t solely about technology—strategic, organizational, and cultural factors are equally critical.

We’ve heard from customers that it’s crucial to consider how prepared your organization is for this technological leap. In this blog post, we’ll share some learnings to help you gauge your AI readiness so you can plan how to move forward effectively.

AI Readiness Wizard

Assess your AI readiness

Get started

Assess your AI readiness

When I meet with customers, they often share that they’re not sure where to start when it comes to assessing their readiness for large-scale AI transformation. It requires a strategic approach to understand your current capabilities, identify areas for improvement, and align these efforts with your business priorities to focus on the areas that will deliver the highest value. We’ve created a new AI Readiness Wizard to help you get started in evaluating your preparedness. Use the assessment to:

• Evaluate your current state. The assessment includes questions that help you determine how well your AI objectives align with your business priorities and the effectiveness of your current data access and security measures. Understanding your starting point is essential for identifying the right next steps.
• Identify gaps. By scoring your responses in the assessment, you can identify focus areas that may need more attention, such as business strategy, AI governance principles, or team expertise. This step helps prepare you for formulating a clear path forward and addressing specific areas for improvement.
• Plan your next steps. Based on your scores, the assessment categorizes your readiness into one of five stages: exploring, planning, implementing, scaling, or realizing. Each stage represents a different level of AI maturity and preparedness, guiding you on where to focus your efforts:
  • Exploring—At this stage, focus on building your AI strategy and experience. You might want to learn about key AI concepts and explore how AI is transforming the business landscape.
  • Planning—Here, you’ll concentrate on formalizing your business strategy. Look at the ways other organizations are driving value with AI and develop an informed plan for prioritizing AI projects.
  • Implementing—This stage involves focusing on leadership support and scaling AI expertise. Ensure you have the necessary resources and expertise to execute your AI initiatives effectively.
  • Scaling—At this level, you’ll aim to create an organization and culture of innovation. Scale your AI initiatives and begin analyzing the impact of AI in your organization.
  • Realizing—Focus on fostering continuous innovation within every team and the organization. Aim to embed AI technology in your operations and culture for sustained value creation.

Assessing your AI readiness requires a strategic approach to understand your current capabilities and identify areas for improvement.”

This assessment offers a structured way to reflect on your current practices and identify key areas to focus on as you develop your strategy for the future. You’ll also find resources for each stage to help you advance.

How AI is reshaping industries

Financial services

We’re seeing the financial services sector make rapid advancements in AI readiness, with 40% of organizations currently in the “implementing” stage. According to recent research, 70% of financial services organizations are using big data analytics in their operations, and 27% have piloted AI applications or AI-assisted solutions.

Additionally, more than half are allocating budgets for AI projects, providing AI-specific training, and fostering internal knowledge sharing. This commitment has enabled 27% of firms to reach the “scaling” and “realizing” stages, surpassing the 25% industry benchmark.

Healthcare

The healthcare industry shows a diverse mix of AI readiness, with 28% of organizations in the “scaling” and “realizing” stages, according to one study. Notably, 44% are actively laying the groundwork in the “exploring” and “planning” stages, focusing on learning and developing their AI strategies. The sector leads in overall maturity, but 14% of organizations report receiving no discernible value from AI, highlighting challenges in measuring the impact of AI investments within their broader business strategies.

Manufacturing

With 38% of organizations in the manufacturing industry still in the “exploring” and “planning” stages, many are focused on learning and developing AI strategies. Research shows that manufacturers actively deploy AI across operations, research and development (R&D), and supply chain management to address key business challenges. 25% believe they achieve significant value from AI implementation.

Manufacturing organizations also have a greater likelihood of appointing AI leadership, which we’re learning enables them to excel in fostering the operational and cultural factors that support value creation, resulting in more firms reaching the “realizing” and “scaling” stages.

Retail

We’ve seen a wide range of AI readiness in the retail sector. While some retailers use AI to enhance customer relationships and drive revenue, research shows that 43% are still in the “exploring” and “planning” stages. This divide is evident between those who adopted cloud technology early—about 25%—and those who have yet to embrace it, with 8% still not using cloud services. Notably, 21% of retailers have a chief AI officer, highlighting commitment among leadership to embed AI into their operations.

Map your AI journey

After assessing your readiness and gathering insights, you’ll want to outline a plan to address gaps and advance through the stages of AI maturity. Our findings at Microsoft have shown that crafting a strategic plan outlining how AI technology will fit into your organizational framework is a great place to start. The AI Strategy Roadmap: Navigating the stages of value creation is a valuable resource designed to guide you through this process.

As you define your AI strategy and roadmap, you might find our e-book, Building a Foundation for AI Success: A Leader’s Guide, helpful in identifying key focus areas for AI implementation.

The post A strategic approach to assessing your AI readiness appeared first on The Microsoft Cloud Blog.

Minimize Risk and Reap the Benefits of AI

Addressing security concerns and implementing safeguards

Get the strategy paper

According to a recent survey conducted by ISMG, the top concerns for both business executives and security leaders on using generative AI in their organization range, from data security and governance, transparency and accountability to regulatory compliance.¹ In this paper, the first in a series on AI compliance, governance, and safety from the Microsoft Security team, we provide business and technical leaders with an overview of potential security risks when deploying generative AI, along with insights into recommended safeguards and approaches to adopt the technology responsibly and effectively.

Learn how to deploy generative AI securely and responsibly

In the paper, we explore five critical areas to help ensure the responsible and effective deployment of generative AI: data security, managing hallucinations and overreliance, addressing biases, legal and regulatory compliance, and defending against threat actors. Each section provides essential insights and practical strategies for navigating these challenges. 

Data security

Data security is a top concern for business and cybersecurity leaders. Specific worries include data leakage, over-permissioned data, and improper internal sharing. Traditional methods like applying data permissions and lifecycle management can enhance security. 

Managing hallucinations and overreliance

Generative AI hallucinations can lead to inaccurate data and flawed decisions. We explore techniques to help ensure AI output accuracy and minimize overreliance risks, including grounding data on trusted sources and using AI red teaming. 

Defending against threat actors

Threat actors use AI for cyberattacks, making safeguards essential. We cover protecting against malicious model instructions, AI system jailbreaks, and AI-driven attacks, emphasizing authentication measures and insider risk programs. 

Grow Your Business with AI You Can Trust

Get the e-book

Addressing biases

Reducing bias is crucial to help ensure fair AI use. We discuss methods to identify and mitigate biases from training data and generative systems, emphasizing the role of ethics committees and diversity practices.

Legal and regulatory compliance

Navigating AI regulations is challenging due to unclear guidelines and global disparities. We offer best practices for aligning AI initiatives with legal and ethical standards, including establishing ethics committees and leveraging frameworks like the NIST AI Risk Management Framework.

Explore concrete actions for the future

As your organization adopts generative AI, it’s critical to implement responsible AI principles—including fairness, reliability, safety, privacy, inclusiveness, transparency, and accountability. In this paper, we provide an effective approach that uses the “map, measure, and manage” framework as a guide; as well as explore the importance of experimentation, efficiency, and continuous improvement in your AI deployment.

I’m excited to launch this series on AI compliance, governance, and safety with a strategy paper on minimizing risk and enabling your organization to reap the benefits of generative AI. We hope this series serves as a guide to unlock the full potential of generative AI while ensuring security, compliance, and ethical use—and trust the guidance will empower your organization with the knowledge and tools needed to thrive in this new era for business.

Additional resources

• Get the Grow Your Business with AI You Can Trust e-book.
• Explore the Introduction to Generative AI and Safety guide.

Get more insights from Bret Arsenault on emerging security challenges from his Microsoft Security blogs covering topics like next generation built-in security, insider risk management, managing hybrid work, and more.

^{1, 2} ISMG’s First annual generative AI study – Business rewards vs. security risks: Research report, ISMG.

The post More value, less risk: How to implement generative AI across the organization securely and responsibly appeared first on The Microsoft Cloud Blog.

Every technology can be used for good or bad. This was as true for fire and for writing as it is for search engines and for social networks, and it is very much true for AI. You can probably think of many ways that these latter two have helped and harmed in your own life—and you can probably think of the ways they’ve harmed more easily, because those stick out in our minds, while the countless ways they helped (finding your doctor, navigating to their office, the friends you made, the jobs you got) fade into the background of life. You’re not wrong to think this: when a technology is new it’s unfamiliar, and every aspect of it attracts our attention—how often do you get astounded by the existence of writing nowadays?—and when it doesn’t work, or gets misused, it attracts our attention a lot.

The job of the people who build technologies is to make them as good as possible at helping, and as bad as possible at harming. That’s what my job is: as CVP and Deputy CISO of AI Safety and Security at Microsoft, I have the rare privilege of leading a team whose job is to look at every aspect of every AI system we build, and figure out ways to make them safer and more effective. We use the word “safety” very intentionally, because our work isn’t just about security, or privacy, or abuse; our scope is simply “if it involves AI, and someone or something could get hurt.”

But the thing about tools is that no matter how safe you make them, they can go wrong and they can be misused, and if AI is going to be a major part of our lives—which it almost certainly is—then we all need to learn how to understand it, how to think about it, and how to keep ourselves safe both with and from it. So as part of Cybersecurity Awareness Month, we’ve created some resources like the Be Cybersmart Kit to help individuals and organizations learn about some of the most important risks and how to protect themselves.

Cybersecurity awareness

Explore cybersecurity awareness resources and training

Get started

I’d like to focus on the three risks that are most likely to affect you directly as individuals and organizations in the near future: overreliance, deepfakes, and manipulation. The most important lesson is that AI safety is about a lot more than how it’s built—it’s about the ways we use it.

Overreliance on AI

Because my job has “security” in the title, when people ask me about the number one risk from AI they often expect me to talk about sophisticated cyberattacks. But the reality is that the number one way in which people get hurt by AI is by not knowing when (not) to trust it. If you were around in the late 1990s or early 2000s, you might remember a similar problem with search engines: people were worried that if people saw something on the Internet, all nicely written and formatted, they would assume whatever they read was true—and unfortunately, this worry was well-founded. This might seem ridiculous to us with twenty years of additional experience with the Internet; didn’t people know that the Internet was written by people? Had they ever met people? But at the time, very few people ever encountered professionally-formatted text with clean layouts that wasn’t the result of a lengthy editorial process; our instincts for what “looked reputable” were wrong. Today’s AI has a similar concern because it communicates with you, and we aren’t used to things that speak to us in natural language not understanding basic things about our lives.

We call this problem “overreliance,” and it comes in four basic shapes:

• Naive overreliance happens when users simply don’t realize that just because responses from AI sound intelligent and well-reasoned, that doesn’t mean the responses actually are smart. They treat the AI like an expert instead of like a helpful, but sometimes naive, assistant.
• Rushed overreliance happens when people know they need to check, but they just don’t have time to—maybe they’re in a fast-paced environment, or they have too many things to check one by one, or they’ve just gotten used to clicking “accept.”
• Forced overreliance is what happens when users can’t check, even if they want to; think of an AI helping a non-programmer write a complex website (are you going to check the code for bugs?) or vision augmentation for the blind.
• Motivated overreliance is maybe the sneakiest: it happens when users have an answer they want to get, and keep asking around (or rephrasing the question, or looking at different information) until they get it.

In each case, the problem with overreliance is that it undermines the human role in oversight, validation, and judgment, which is crucial in preventing AI mistakes from leading to negative outcomes.

How to stay safe

The most important thing you can do to protect yourself is to understand that AI systems aren’t the infallible computers of science fiction. The best way to think of them is as earnest, smart, junior colleagues—excited to help and sometimes really smart but sometimes also really dumb. In fact, this rule applies to a lot more than just overreliance: we’ve found that asking “how would I make this safe if it were a person instead of an AI?” is one of the most reliable ways to secure an AI system against a huge range of risks.

1. Treat AI as a tool, not a decision-maker: Always verify the AI’s output, especially in critical areas. You wouldn’t hand a key task to a new hire and assume what they did is perfect; treat AI the same way. Whether it’s generating code or producing a report, review it carefully before relying on it.
2. Maintain human oversight: Think of this as building a business process. If you’re going to be using an AI to help make decisions, who is going to cross-check that? Will someone be overseeing the results for compliance, maybe, or doing a final editorial pass? This is especially true in high-stakes or regulated environments where errors could have serious consequences.
3. Use AI for brainstorming: AI is at its best when you ask it to lean into its creativity. It’s especially good at helping come up with ideas and interactively brainstorming. Don’t ask AI to do the job for you; ask AI to come up with an idea for your next step, think about it and maybe tweak it a bit, then ask it about its thoughts for what to do next. This way its creativity is boosting yours, while your eye is still on whether the result is what you want.

Train your team to know that AI can make mistakes. When people understand AI’s limitations, they’re less likely to trust it blindly.

Impersonation using AI

Deepfakes are highly realistic images, recordings, and videos created by AI. They’re called “fakes” when they’re used for deceptive purposes—and both this threat and the next one are about deception. Impersonation is when someone uses a deepfake to convince you that you’re talking to someone that you aren’t. This threat can have serious implications for businesses, as bad actors can use deepfake technology to deceive others into making decisions based on fraudulent information.

Imagine someone creates a deepfake of your chief finance officer’s voice and uses it to convince an employee to authorize a fraudulent transfer. This isn’t hypothetical—it already happened. A company in Hong Kong was taken for $25.6 million with the use of this exact technique.¹

The real danger lies in how convincingly these AI-generated voices and videos can mimic trusted individuals, making it hard to know who you’re talking to. Traditional methods of identifying people—like hearing their voice on the phone or seeing them on a video call—are no longer reliable.

How to stay safe

As deepfakes become more compelling, the best defense is to communicate with people in ways where recognizing their face or voice isn’t the only thing you’re relying on. That means using authenticated communication channels like Microsoft Teams or email rather than phone calls or SMS, which are trivial to fake. Within those channels, you need to check that you’re talking to the person you think you’re talking to, and that software (if built right) can help you do that.

In the Hong Kong example above, the bad actor sent an email from a fake but realistic-looking email address inviting the victim to a Zoom meeting on an attacker-controlled but realistically-named server, where they had a conversation with “coworkers” who were actually all deepfakes. Email services such as Outlook can prevent situations like this by vividly highlighting that this is a message from an unfamiliar email address and one that isn’t part of your company; enterprise video conferencing (VC) systems like Teams can identify that you’re connecting to a system outside your own company as a guest. Use tools that provide indicators like these and pay attention to them.

If you find that you need to talk over an unauthenticated channel—say, you get a phone call from a family member in a bad situation and desperately needing you to send them money, or you get a WhatsApp message from an unfamiliar number—consider pre-arranging some secret code words with people you know so you can identify that they’re really who they say they are.

All of these are examples of a familiar technique that we use in security called multi-factor authentication (MFA), which is about using multiple means to verify someone is who they say they are. If you communicate over an authenticated channel, an attacker has to both compromise an account on your service (which itself should be protected by multiple factors) and create a convincing deepfake of that particular person. Forcing attackers to simultaneously do multiple different attacks against the same target at once makes the job exponentially harder for them. Most important services you use (email, social networks, and so on) allow you to set up MFA, and you should always do this when you can—preferably using “strong” MFA methods like physical keys or mobile apps, rather than weak methods like SMS, which are easily faked. According to our latest Microsoft Digital Defense Report, implementing modern day MFA reduces the likelihood of account compromise by 99.2%, significantly strengthening security and making unauthorized access more difficult for attackers to gain access. Although MFA techniques reduce the risk of identity compromise, many organization have been slow to adopt them. So, in January 2020, Microsoft introduced “security defaults” that turn on MFA while turning off basic and legacy authentication for new tenants and those with simple environments. The impact is clear: tenants that use security defaults experience 80% fewer compromises than tenants that don’t.

Scams, phishing, and social manipulation

Beyond impersonating someone you know, AI can be used to power a whole range of attacks against people. The most expensive part of running a scam is taking the victim from the moment they first pick up the bait—answering an email message, perhaps—to the moment the scammers get what they want, be it your password or your money. Phishing campaigns often require work to create cloned websites to steal your credentials. Spear-phishing requires crafting a targeted set of lures for each potential victim. All of these are things that bad actors can do much more quickly and easily with AI tools to help them; they are, after all, the same tools that good actors use to automate customer service, website building, or document creation.

On top of scams, an increasingly important use of AI is in social manipulation, especially by actors with political goals—whether they be real advocacy organizations or foreign intelligence services. Since the mid-2010s, a key goal of many governments has been to sow confusion in the information world in order to sway political outcomes. This can include:

• Convincing you that something is true when it isn’t—maybe that some kind of crime is rampant and you need to be protected from it, or that your political enemies have been doing something awful.
• Convincing you that something isn’t true when it is—maybe that the bad things they were caught doing are actually deepfakes and frauds.
• Simply convincing you that you can’t know what’s true, and you can’t do anything about it anyway, so you should just give up and stay home and not try to affect things.

There are a lot of tricks to doing this, but the most important ones are to make it feel like “everybody feels” something (by making sure you see just enough comments saying something that you figure it must be right, and you start repeating them, making other people believe it even more) and by telling you what you want to hear—creating false stories that line up with what you’re already expecting to believe. (Remember motivated overreliance? This is the same thing!)

AI is supercharging this space as well; it used to be that if you wanted to make sure that every hot conversation about a subject had people voicing your opinion, you needed either very non-human-sounding scripts, or you needed to hire a room full of operators. Today, all you need is a computer.

You can learn more about these attacks in on our threat intelligence website called Microsoft Security Insider.

How to stay safe

Take your current habits for being aware of potential scams or phishing attempts, and turn them up a notch. Just because something showed up at the top of search results doesn’t mean it’s legitimate. Look at things like URLs and source email addresses carefully, and see if you’re looking at something genuine or not.

To detect sophisticated phishing attempts, always verify both the source and the information with trusted channels. Cybercriminals often create a false sense of urgency, use amplification tactics, and mimic trustworthy sources to make their emails or content appear legitimate. Stay especially cautious when approached by unfamiliar individuals online, as most fraud or influence operations begin with a simple social media reply or a seemingly innocent “wrong number” message. (More sophisticated attacks will send friend requests to people, and once you get one person to say yes, your further requests to their friends will look more legitimate, since they now have mutual “friends” with the attacker.)

Social manipulation can affect you both directly (you see messages created by a threat actor) or indirectly (your friends saw those messages and unwittingly repeated them). This means that just because you hear something from someone you trust, you can’t be sure they didn’t get fooled too. If you’re forming your opinion about something, or if you need to make an important decision about whether you believe something or not, do some research, and figure out where a story came from. (And don’t forget that “they won’t tell you about this!” is a common thing to add to frauds, just to make you believe that the lack of news coverage makes it more true.)

But on the other hand, don’t refuse to believe anything you hear, because making you not believe true things is another way you can be cheated. Too much skepticism can get you in just as much trouble as not enough.

And ultimately, remember—social media and similar fora are designed to get you more engaged, activated, and excited, and when you’re in that state, you’re more likely to amplify any feelings you encounter. Often the best thing you can do is simply disconnect for a while and take a breather.

The power and limitations of AI

While AI is a powerful tool, its safety and effectiveness rely on more than just the technology itself. AI functions as one part of a larger, interconnected system that includes human oversight, business processes, and societal context. Navigating the risks—whether it’s overreliance, impersonation, cyberattacks, and social manipulation—requires not only understanding AI’s role but also the actions people must take to stay safe. As AI continues to evolve, staying safe means remaining active participants—adapting, learning, and taking intentional steps to protect both the technology and ourselves. We encourage you to use the resources on the cybersecurity awareness page and help educate your organization so as to create a security-first culture and secure our world—together.

Learn more about AI safety and security

• Take the next step in developing cybersecurity skills and awareness.
• Learn more about Trustworthy AI.

¹Finance worker pays out $25 million after video call with deepfake ‘chief financial officer’, CNN, 2024.

The post AI safety first: Protecting your business and empowering your people appeared first on The Microsoft Cloud Blog.

When businesses infuse cutting-edge innovations like AI into their operations, it can drive sustainable, long-term growth, futureproof them against economic headwinds, and create lasting competitive advantage. While the market dialog is dominated by incredible new AI-driven services, there is untapped value in the hundreds of millions of existing applications that can now be modernized and infused with AI.

Without modernization, organizations may miss out on the full value of their investments, lag behind the competition, or fall prey to costly disruptions—and they certainly won’t be positioned to develop tomorrow’s leading AI innovations. Pitfalls like these are already a reality for many: according to the Forrester study exploring modernization, one in four business decision-makers experienced digital platform failures due to modernization challenges.

Because true modernization can (and should) touch every area of your business, it gets complex, involving everything from processes and systems to orchestration and strategic planning. As a result, many leaders aren’t sure where to begin—which is why we’ve created this blog series to chart a course through modernization with AI readiness in mind. 

In this first blog, I’ll cover what modernization really means, why it matters for your business, and how to think about modernization as your path to AI value.

Modernization is a holistic approach 

The first step on your modernization journey is to get clear about what differentiates modernization from more piecemeal approaches. Modernization means updating and improving assets across all business areas so they work well with evolving digital software and the cloud. This is a holistic approach that includes people, processes, and skillsets along with data, apps, and infrastructure.

Modernization breathes new life into legacy technologies to prime them for AI

Earlier I mentioned that modernization is critical to the future of your business—let’s give a little background as to why. For decades, businesses grew up alongside purpose-built digital solutions that met the needs of the day. These solutions are now struggling to meet current needs. For example, they can’t easily incorporate the latest AI capabilities because they weren’t built for fast-paced innovation cycles. This tech also often works in siloes and may not be able to process large amounts of data to support the latest intelligent services. Take it from Sapiens, an insurance platform provider across thirty countries: they struggled to innovate because their digital practices were established before they adopted the cloud. 

Modernization helped Sapiens overcome these issues and prime their operations for AI innovation by migrating, transforming, and distributing their key applications. This allowed them to accelerate their development processes and their innovation cycle, since developers could devote more time and resources to improvements rather than maintenance.

“Using and investing in Microsoft Azure tools to automate some of our infrastructure and processes, we managed to cut our time to market in half and reduced our operational overhead by at least 40 percent.”—Michael Mirel, Head of Cloud and DevOps Center of Excellence at Sapiens.

They’re not alone—according to an IDC survey on the benefits of cloud migration and modernization, 41% of organizations cite operational efficiency and 30% cite cost savings as the top outcomes achieved. When businesses take this technology posture, they lay the groundwork for AI and analytics innovation and more agile operations. 

Modernization starts with the cloud, but it doesn’t end there 

Migrating to the cloud is essential (30% of respondents in that same IDC survey reported the cloud eased modernization), but think of the cloud as just one aspect of driving scale, achieving long-term outcomes, and unlocking the potential of technologies like AI. 

Take Scandinavian Airline Systems (SAS) as a case study. They incorporated the cloud as part of SAS Forward, a broad strategy to help adapt to changing market dynamics and cost pressures.

“The airline industry is highly competitive, to continue to provide great experiences for travelers, we needed to make big changes.”—Mikael Perhult, Tech Lead, Cloud at SAS – Scandinavian Airlines.

That’s why SAS Forward went beyond cloud migration, modernizing SAS’s databases and apps as a fully managed platform service on Azure.

“We wanted to transform the technologies that support and connect SAS’ systems and services for greater scalability, efficiency, and security, while paving the way for innovation for our customers.”—Prakash Ujjwal, Senior Systems Specialist, IT Infrastructure Services. 

Just as SAS went beyond initial migration, they also had to go beyond the tech itself. This meant a thoughtful reimagining of how their operations could be modernized to maximize investment.

Modernization reshapes more than technology 

While apps and platforms are essential to modernization, businesses should also align their people, processes, and skillsets to ensure the entire enterprise is working toward the same goal. This is especially true with AI, since it creates a new way of working that disrupts long-standing routines. In other words, AI is both a catalyst for modernization and accelerator of modernization. 

The research behind the Forrester Application Modernization Checklist supports this idea. One in five decision-makers reported achieving modernization, and those who succeeded said they did so because they looked at the task holistically. They built a clear strategy around AI and AI training, tied AI closely to business outcomes, and used cross-org metrics to track and measure progress and impact. And crucially, they drew on support from their strategic partners to upgrade their technology stacks—modernization is too important and too complex to tackle alone. SAS embodied this holistic approach by setting goals to transform operations and maintenance, increase overall innovation, and leverage partners to get more out of their initial cloud investments. They succeeded and streamlined developer workflows to enable continuous improvement while modernizing airline operations at scale.

The right modernization approach unlocks competitive advantage 

When businesses take a truly holistic approach to modernization, transformative outcomes speak for themselves. Sapiens reports that they’re able to extract deeper insights more easily and drive more effective business decisions, while SAS built an environment that fosters long-term innovation and has improved customer experiences.

I invite you to continue charting a modernization course with me in the next blog in our series. We’ll walk through specific steps for effective modernization identified by the Forrester Consulting Application Modernization Checklist that Microsoft commissioned.

• Download the 2024 commissioned Forrester Consulting study: Unlock Competitive Advantage with Application Modernization. 

• Read the accompanying Application Modernization Checklist. 

• Learn about the benefits of cloud migration and modernization from IDC. 

• Consult the Microsoft Application and Data Modernization: Azure page. 

• Visit the Microsoft Azure Innovate site. 

The post Achieving AI readiness through comprehensive modernization appeared first on The Microsoft Cloud Blog.

Large language models (LLMs) are AI systems based on transformer architectures and trained on vast amounts of text data to understand and generate human-like text. Using deep learning techniques, LLMs process and produce accurate responses rapidly. Deep learning is a subset of machine learning that uses multi-layered neural networks to simulate the complex decision-making power of the human brain.

Large language models are trained on a massive volume of data, and once properly trained, they have a broad applicability for a range of natural language processing and machine learning applications. LLMs are typically multiple billions of parameters in size, making them five to ten times larger than small language models (SLMs).

Azure AI Services

Build cutting-edge, market ready AI applications

Try for free

What can LLMs do?

Large language models (LLMs) offer significant benefits across various industries by automating and enhancing numerous tasks involving natural language processing. These AI-powered tools can rapidly analyze vast amounts of text data, generate human-like content, and provide intelligent responses to queries. However, always keep in mind that any content created by AI models and used in final deliverables must not infringe on copyrights or intellectual property rights of the original owners.

• In business, LLMs may improve customer service through chatbots, streamline document analysis, and assist with market research.
• In healthcare, LLMs may assist clinicians with reviewing medical literature and clinical documentation.
• In education, LLMs may help teachers create personalized learning materials and provide instant tutoring assistance for their students.
• In the legal industry, LLMs may help law firms with contract analysis and legal research.

Additionally, LLMs can help support content ideation for marketing, journalism, and creative industries.

Let’s take a brief tour through the world of large language models.

5 key features and benefits of LLMs

While there are many benefits of large language models, here are five to consider:

1. Natural language understanding

The model can interpret context, detect sentiment, and understand idiomatic expressions and colloquialisms. It can often infer unstated information and respond appropriately to ambiguous queries. Also, LLMs can combine information from various sources in their training data to answer complex questions, solve problems creatively, translate languages, and even assist in research and innovation.

Benefit: LLMs can comprehend context, nuance, and intent in the text that was input into it, which allows for more intuitive human-computer interaction. The large language model enables the discovery of new insights and connections across diverse fields. It also powers more intelligent search engines that provide direct, human-like answers to queries rather than just links to relevant pages.

2. Versatile multimodal generation

LLMs can produce coherent and contextually appropriate outputs in multiple styles, languages, and formats—from poems and stories to emails, technical reports, and even spoken language. With advancements in multimodality, these models now extend beyond text to support speech, images, and other forms of media. This facilitates global communication, broadens access to information, performs translation tasks, question-answering, generating code with minimal additional training, and even understanding code-switching within conversation or between different media types.

Benefit: Synthesizing knowledge across text, speech, and other modalities saves time and resources in content creation across various domains. The models can analyze and determine sentiment or emotional tone in both text and speech, which is valuable for market research, customer feedback reviews, and even personalized interactions like voice-based assistants or multimedia content generation.

3. Code generation and analysis

Large language models can produce code as well as text. For example, LLMs can assist developers by generating code snippets, functions, or even entire programs based on natural language descriptions. They can also analyze existing codebases to help identify bugs, suggest optimizations, and provide explanations of complex code sections, effectively serving as an AI-powered coding assistant. In addition, LLMs can assist developers with:

• Building applications
• Auto-completing code
• Finding errors in code
• Analyzing and debugging software code
• Offering round-the-clock assistance without fatigue
• Creating test cases based on function specifications
• Creating entire code blocks in various programming languages
• Suggesting appropriate design patterns for given problems
• Suggesting improvements for code readability and maintainability
• Identifying security issues across multiple programming languages

Benefit: Developers can tailor the code to specific industries and use cases, thus adapting the model to specialized domains like healthcare, law, marketing, customer service, scientific research, and finance.

4. Task-specific without fine-tuning

With their massive knowledge base, LLMs can perform tasks such as summarization, translation, question-answering, and code generation with minimal additional training. The LLMs can be retrained periodically to respond in a more human-like manner, incorporate new data, and improve performance. 

Benefit: Reduces the need for specialized models for different tasks since they are so capable. LLMs excel at generating content that sounds natural, across multiple subject areas, with high accuracy.

5. Scalability and efficiency

LLMs can process long-form content or analyze extensive documents in parallel, leveraging graphics processing unit (GPU) capabilities for faster training and inference. This allows for efficient handling of large-scale language tasks and rapid generation of responses.

Benefit: Easily handles increased workloads and adapts to growing business needs. They can analyze large volumes of text data to extract insights and patterns, aiding in decision-making processes and boosting productivity.

Use LLMs to build comprehensive AI solutions to revolutionize industries

LLMs have revolutionized natural language processing by offering robust capabilities for understanding and generating human-like text. Despite their significant advancements, there are still some limitations. To ensure their ethical and appropriate use across various sectors, continuous improvements are necessary as we move forward.

LLMs can be used with other Microsoft Azure AI products to build advanced and comprehensive solutions to suit most industries. Their features and benefits make them an attractive option for businesses seeking to enhance natural language processing capabilities across various applications—from customer service to content creation and software development.

The ability of large language models to understand context, generate coherent text, and adapt to specific domains makes them versatile and valuable tools that are not only applicable in fields beyond just language processing—such as software development, data science, decision support systems, and creative industries—but that organizations can rely on to boost productivity, efficiency, and innovation across sectors.

Introduction to large language models

Learn how to use LLMs to improve your workflow

Get started

Our commitment to responsible AI

Organizations across industries are leveraging Azure OpenAI Service and Microsoft Copilot services and capabilities to drive growth, increase productivity, and create value-added experiences. From advancing medical breakthroughs to streamlining manufacturing operations, our customers trust that their data is protected by robust privacy protections and data governance practices. As our customers continue to expand their use of our AI solutions, they can be confident that their valuable data is safeguarded by industry-leading data governance and privacy practices in the most trusted cloud on the market today. 

At Microsoft, we have a long-standing practice of protecting our customers’ information. Our approach to responsible AI is built on a foundation of privacy, and we remain dedicated to upholding core values of privacy, security, and safety in all our generative AI products and solutions.

Get started with Azure OpenAI Service

• Explore Azure Open AI Service.
• Watch this video to learn more about the Azure AI model catalog.
• Read about how you can Deploy large language models responsibly with Azure AI.

Learn more about AI solutions from Microsoft

• Explore Microsoft AI solutions to fuel your AI transformation.
• Learn how to build and optimize your strategic plan for AI with the AI Strategy Roadmap.
• Read An Introduction to Generative AI and Safety.

The post 5 key features and benefits of large language models appeared first on The Microsoft Cloud Blog.