Build advanced, reliable automations faster with AI 

Developers are key to business process innovation. In Power Automate, it is now possible to build automations that utilize AI within cloud flows using generative actions, easily use natural language to create expressions for flows with Copilot for expressions, and incorporate document processing into workflows with next-generation document processing. These features enhance developers’ ability to deliver solutions efficiently and effectively. 

Introducing next-generation document processing with GPT 

Efficient document management is vital for smooth operations and timely decisions, but traditional AI models require lengthy training and frequent updates. With GPT-4-powered document processing, now available in public preview, you can use natural language to extract and process unstructured data from documents like invoices, medical results, or emails. This next-gen tool handles multi-modal content, automating email, document, and image processing for improved productivity. For makers who want to blend in pro-code or gain greater control, we’ve added the ability to structure outputs using JavaScript Object Notation (JSON) directly inline. Makers can combine multi-modal content with enterprise knowledge from Microsoft Dataverse in their prompts. This feature enables makers to quickly deploy document and image processing directly into existing workflows using natural language instructions.  

Learn more in our detailed blog Boost productivity: Automate emails, documents, and images with Microsoft AI Builder.
Visit our Use AI Builder in Power Automate documentation
Join our Power Automate session for a deep dive. 

Transform your automation processes with generative actions 

Generative actions allow you to add AI-powered steps to your cloud flows, letting AI reason over enterprise guardrails provided in unstructured content format such as documents and PDFs, and decide the best course of action for complex tasks. For example, with generative actions you can automate conditional approvals based on the content of incoming emails and company policies. This feature, currently in the Early Access Program, is coming soon to public preview, when you will be able to create, edit, test, and run generative actions all from within the cloud flow designer. 

Create and edit expressions faster with Copilot 

Copilot for expressions, now in public preview, is designed to assist developers and makers in creating or editing complex expressions using natural language. By simply describing what you want to build and referencing the dynamic data in the flow, Copilot translates your prompts into specific expressions in your Power Automate flow, speeding up development and reducing manual coding time. 

To learn more about this feature and how to get started, read our Create and edit expressions with Copilot documentation.

Streamline automation and operations  

Monitor and manage at scale with new automation center 

Automation center, now generally available, is a central hub for efficient monitoring and troubleshooting experiences for automation processes across Power Automate at scale. The automation center provides comprehensive visualizations that enable you to monitor the health of your automations, quickly detect issues or trends, and troubleshoot problems more efficiently. 

Whether you’re a developer, operator, Center of Excellence team member, or business analyst, the automation center provides a centralized view of all automation activities. Now generally available, it features a user-friendly interface with dashboards that show recommendations, the health status of work queues, and desktop flow activity. Additionally, new views for execution logs, performance metrics, and integrated agents are now available in public preview. 

To learn more and to get started, read our Automation center documentation.

Improve robotic process automation (RPA) reliability using repair with Copilot 

Repair with Copilot, now in public preview, simplifies resolving user interface (UI) control issues in both attended and unattended RPA with Power Automate for desktop. By proactively providing solutions to common automation issues, such as notifying the right person in the automation center, repair with Copilot reduces downtime for RPA scenarios.  

Example of using repair with Copilot for unattended RPA:

Example of using repair with Copilot for attended RPA: 

To learn more and get started with repair with Copilot, read our UI and browser automation actions documentation.

Start your journey into a new era of automation and AI 

Are you ready to bring your organization into the future with the latest AI capabilities in Power Automate? Get started with a free trial or compare plans, join our sessions below at Microsoft Ignite 2024, and read about other exciting announcements for Microsoft Power Platform.  

Power Automate

Find the right low-code automation plan for your business.

Learn more

• Revolutionize the way you work with Power Automate and AI with Joe Fernandez and Heather Orta-Olmo  
• Automate e-mail, doc, and image processing with GPT in Power Automate with Gwenael Bego and Maria Gutierrez 
• Unlock innovation with practical strategies in Power Automate with Jonathan Kendall 

The post The next frontier in enterprise intelligent automations with Microsoft Power Automate appeared first on Microsoft Power Platform Blog.

What are Prompts? 

Prompts are the instructions that we can give to large language models (LLMs) like GPT to perform specific tasks. The more precise the instructions we give, the better GPT performs at executing the tasks. To allow companies to create effective prompts we’ve created a new tool, Prompt Builder, a low-code prompt engineering interface that is seamlessly integrated in Power Automate. 
Customers are already excited about the perspectives that Generative AI can bring to Power Automate. Epiq, a global leader in technology-enabled services, has set a new standard with its innovative use of Generative AI by applying it to transform their HR operations. “With AI Builder in Power Automate, we added a model that cleans up formatting of new employee submission forms. This allows us to process more efficiently at scale – and it works flawlessly.” says Colt Coan, Senior Manager of Rapid Development, Epiq. Want le learn more? Read Epiq’s customer story article.

How to build a Prompt 

To access Prompt Builder, start by adding the “Create text with GPT using a prompt” action from the AI Builder connector:  

You’ll then be able to use Prompt Builder, AI Builder’s new low-code prompt engineering interface, to create your instructions to GPT that will be executed during the workflow. 
We’ve added examples of Prompts to help you get started. Common scenarios include: 

• Classify Text: E-mail classification and case routing. 
• Summarize Text: Summarization of e-mails, contracts or meeting notes. 
• Extract action points from text: Identify upsell/cross-sell opportunities and next best actions. 
• Extract information from text: Understand customer requests to automate sales quotes and proposals. 
• Respond to a complaint: Draft e-mail responses to fast-track support case management. 

When creating a prompt, you’ll be able to add dynamic values in the instructions so that the Prompt can automatically be executed over new data, for example an incoming e-mail. 


In the body of the prompt, you’ll be able to craft your GPT instructions, add multiple dynamic values, define the output format you want, and test your prompt with sample data:

Once you’re happy with the way your prompt performs, you can save your prompt for future use. You’ll be able to access all your saved Prompts along with a library of pre-built Prompts from the new AI Prompts page. Prompts can be shared with other users, added to solutions for cross-environment deployment, and invoked across other products like Power Apps or Copilot Studio. To learn more about how to create effective Prompts, refer to our new Prompt engineering guide.

Human in the loop validation 

Content generated by AI can include mistakes and surprises, so it’s important to make sure it is accurate and appropriate before using it. To facilitate this process, you can include GPT-generated text in an approval flow that allows you to validate and edit the content before using it: 

Approval workflows to validate GPT content are highly recommended but no longer mandatory now that the feature is generally available. 

Available in the United States, Australia, UK, and Europe  

GPT Prompts and Prompt Builder will be generally available this week across the United States, Australia, the United Kingdom, and Europe. 
GPT Prompts currently leverage the GPT-3.5-Turbo model from Azure’s OpenAI Service, meaning your prompts and outputs: 

• are NOT available to other customers. 
• are NOT available to OpenAI. 
• are NOT used to improve OpenAI models. 
• are NOT used to improve any Microsoft or 3rd party products or services. 
• are NOT used for automatically improving Azure OpenAI models for your use in your resource
   

Please reference the Azure OpenAI Service data, privacy and security documentation for details on how your data will be processed, used and stored with GPT Prompts. 

Get Started

• Try Prompt Builder: https://aka.ms/tryprompts  
• Prompt Engineering Guide: https://aka.ms/promptguide 
• Documentation: Prompts Overview | Microsoft Learn
• Learning Videos: Overview Video | GPT Prompts with Excel Video 
• Customer Story: Epiq – Revolutionizing HR operations with Power Platform + Generative AI 
• Pricing for GPT Prompts can be found in the Microsoft Power Platform Licensing Guide 
• Governance: Learn about administrating Generative AI features and cross region availability  

The post AI Builder GPT Prompts are generally available appeared first on Microsoft Power Platform Blog.

The company began its business back in 2018, and since then has been using Microsoft Power Platform to automate some complex business processes. Since their early adoption of the platform, Nsure.com has grown their business with more complex processes as well as an increasing number of contacts from their potential and existing customers. Between 2020 and 2023, Nsure’s revenue grew at a Compound Annual Growth Rate (CAGR) of over 100%.

Recently, Nsure.com has been adopting and infusing Power Platform’s generative AI and copilot capabilities into some business processes to support sustainable growth of the business without increased operational costs.

In this story, we’ll learn how Nsure’s business divisions have built automations with Power Platform and generative AI to increase efficiencies. Additionally we’ll learn how Nsure’s IT division, led by CTO Jakub (Kuba) Skalbania and his team, are integrating with other systems, building a Microsoft Fabric-based analytics platform, and using copilot features in Power Platform to rapidly build new automations to streamline internal processes by infusing generative AI – reducing manual processing time by over 60% and cost by 50%.

“The use of Power Platform, including Power Automate, along with generative AI, enables us to achieve hyper-growth without a substantial increase in team size. Nsure.com can markedly enhance client interactions while improving the quality of service. This is particularly crucial in the labor-intensive insurance sector, which is characterized by complex processes for both clients and agents.”
Adrian Dzielnicki, CFA – CEO, Nsure.com

Challenges of a rapidly growing business

Home and auto insurance policy purchasing is a complex business in the United States. It requires insurance brokers to make sure the customer’s policy purchasing process adheres to the requirements of each state. Because Nsure.com provides insurance policies from over 50 insurers, this complexity is multiplied.

With the company quickly growing to nearly 1 million unique users of the platform, the company had to think of a way to sustainably evolve the business, by trying to reduce any mundane and repetitive manual processes.

One of the primary challenges for Nsure.com was the overwhelming human involvement required in customer contacts. Nsure.com was dealing with over 1,000 data points for each unique user: information about the customer’s home, vehicles etc. – equating to over 1 billion datapoints in total. Each month, around 100 customer representatives processes over 100,000 interactions with the customer through calls, emails and texts.

Customer representatives had to manually respond to each of these contacts as well as refer to related records. The manual nature of these tasks not only consumed a lot of time but also increased the risk of errors.

Supporting business processes with generative AI capabilities in Power Platform

Nsure.com and their technology partner Netwise decided to try out the newly released Copilot and generative AI capabilities of Power Automate to streamline Nsure’s insurance processes.

The team focused on inbound communications (emails or text messages). They built Power Automate cloud flows that assess priority, generate tailored summaries, gauge intent, route to relevant departments, and even suggest possible next steps for agents. This ensures that every customer interaction is met with a tailored and efficient response.

With the help of Copilot in cloud flows, Nsure.com was able to construct these flows that seamlessly integrate Dataverse and AI Builder, using the power of natural language processing to build cloud flows from natural language prompts.

“What’s really incredible about one of the flows is that it only took us 40 minutes to create and totally configure using Copilot. Usually this would require an extensive process of dragging-and-dropping and clicking–it would take at least 4 hours.”
Jakub Skalbania, CGO, Netwise, acting CTO at Nsure.com

Following are some detailed examples of how Nsure.com uses generative AI to manage its customer contacts.

Summarization, categorization, and routing customer communications

Nsure’s shared inboxes were flooded with a large number of emails on a daily basis. As a result, customer representatives found it difficult to triage and identify which contacts needed to be worked on next.

To solve this, a Power Automate cloud flow is triggered when an email is received in the shared mailbox. Once an email is processed, generative AI is used to triage and decide whether that email requires immediate attention or not. Sometimes email responses are automatically generated; representatives just need to review the response before sending that email to the customer.

If the email requires manual intervention, generative AI summarizes the contents of the email, compiles, and routes ensures that all pertinent information is channeled to a single Teams group, making it easily accessible to all relevant stakeholders with appropriate priority.

Information extraction from customer emails and text messages

One of the critical components of customer satisfaction is ensuring that details (such as the customer’s available times) are captured when scheduling follow-up calls. These details are not always easy to capture or extract manually and can lead to missing information.

Nsure.com decided to use AI builder to map unstructured data collected by Power Automate flows from email, chat, and call transcriptions and uses these data points to trigger new actions and flows.

For example, if the customer wants to change the appointment time and sends a text message “I have an appointment at 4”, AI builder can extract the fact that this appointment is for today, at 4:00PM as shown in the image below and search for the existing 4:00PM appointment in the system automatically.

This is then integrated with their custom booking system to secure the next available timeslot of the representatives.

By understanding and recording key information the customer has provided, customer satisfaction is increased. The company has observed a significant reduction in customer complaints, resulting in a 4.7 (out of 5) star rating from over 4,000 reviews on review sites.

Data transformation from legacy systems

Customer representatives often needed to access multiple systems to provide accurate answers about policy details, as each state’s requirements are different and insurers provide different policies. Finding this information was one of the most time-consuming and mundane tasks for the customer representatives.

To automate this process, Nsure.com developed a Power Automate desktop flow (RPA) which gathers responses from these systems, which are automatically consolidated and transformed using generative AI models to specific formats that can be presented to customers and also transferred to other systems.

1. Power Automate is triggered when a new processing request is received and added to the Azure Service Bus queue
2. Data is extracted from Azure Blob Storage
3. Power Automate desktop flows are called to run the automation in virtual machine
4. Based on the type of case, relevant sub flow is called
5. Result from the virtual machine is sent back to the Azure Service Bus queue.

Content generation and automated reply suggestions

When customers purchase homeowner’s insurance in the United States, insurance carriers require for some states or counties additional documentation: Wind mitigation reports, 4 point inspections, construction permits. Previously agents had to manually review these documents, extracting the necessary information, and copying it.

With the help of generative AI and Power Automate flows, key information from the documents, can be automatically extracted. Generative AI analyzes the unstructured text from these documents. The outcomes of these analyses are then fed into another flow and model, ensuring that the best policy coverages are determined for the customer based on real-world data. Then, generative AI is applied to analyze returned offers from insurers, compare them, and send the best options to the client. It is a very complicated process, as each of the 50 insurers on the Nsure.com platform has their own product specifics.

Next-best action and email reply suggestions are also automatically generated and combined with other transactional data, activity data, and the communication history of the customer.

With manual tasks reduced by 60%, employees can focus on more valuable tasks

With all of these automations in place, Nsure’s customer service representatives reduced their time on manual tasks by more than 60, allowing them to focus on more valuable and meaningful tasks – engaging with high-priority, complex escalations; building customer relationships; and increasing satisfaction.

Instead of processing every contact, the team can now oversee the communications that are started by AI and make final decisions – where generative AI is essentially providing a copilot-like experience.

Future optimizations planned using Microsoft Power Platform

The flows described above are only the beginning for Nsure.com. In the future, Nsure.com has eliminate mundane and repetitive tasks with generative AI and Copilot capabilities as well as do more with less by having virtual assistants in call campaigns for sales of new policies, upselling, and cross-selling. Nsure.com aims to have more than 90% of incoming requests handled by AI virtual assistants to support requests for insurance quotes and policies – including activities such as submitting payments, reviewing renewal offers, and requesting discounts.

“Building flows with Copilot was insanely easy. I just had to describe it in natural language and it created the flow exactly as I needed it. I just configured the steps from there.”

Jakub Skalbania, CGO, Netwise, acting CTO at Nsure.com

Share your story with us!

Do you have a story or experience to share? We are excited to learn more about how individuals and organizations are using Power Platform with generative AI and Copilot capabilities! Submit your story at https://aka.ms/ShareAIStory

Availability

These features are currently generally available in the United States and English language. Other regions may require you to select United States environments. Note that some of these features are in preview, and accuracy and relevance cannot be guaranteed. 

Responsible AI

Microsoft is committed to creating responsible AI by design. Our work is guided by a core set of principles: fairness, reliability and safety, privacy and security, inclusiveness, transparency, and accountability. We are helping our customers use our AI products responsibly, sharing our learnings, and building trust-based partnerships. For these new services, we provide our customers with information about the intended uses, capabilities, and limitations of our AI platform service, so they have the knowledge necessary to make responsible deployment choices. Responsible AI FAQs are part of a broader effort to put Microsoft’s AI principles into practice. To find out more, see Microsoft AI principles.

The post Insurance agency Nsure.com leverages Microsoft Power Platform and generative AI to reduce manual processes by 60%+ appeared first on Microsoft Power Platform Blog.

GPT action broadly available to US customers

The ‘Create text with GPT action’ is now broadly available to customers based in the United States. This exciting new capability allows makers to use GPT models in the context of automation to perform a wide range of enterprise scenarios centered around text processing and content generation. Available through the AI Builder connector, this capability allows you to send dynamic content to GPT and perform numerous tasks like summarizing content, extracting content, classifying e-mails or pre-generating e-mail responses.

You can create custom instructions to GPT or use pre-configured templates to get started. Learn more about how to use GPT from Azure’s OpenAI service with the AI Builder connector: Use your Azure OpenAI Service model in Power Automate (preview) | Microsoft Learn

Validate and augment GPT content with editable Approvals

An essential part of using AI-generated content in automated workflows is the ability for human review. AI-generated content can be imperfect, and it is important to be able to review it, especially if it is used for customer-facing communication.

For you to include a human review process within GPT flows, we’ve introduced a new feature in Approvals that allows you to review and edit text. Add the new Power Automate action “Start and wait for an approval of text” after your GPT action to send the AI-generated text to an individual approver or a team of approvers.

This action pauses the flow and creates an “Approve/Reject – First to respond” approval with the ability to edit the suggested text generated by GPT.

It has never been easier to review and edit text with the power of Approvals! Learn more about this feature: Use your Azure OpenAI Service model in Power Automate (preview) | Microsoft Learn.

AI Builder activity monitoring available in public preview

We’re happy to introduce a new monitoring feature that allows you to keep track of AI Builder model activity and AI credit consumption.

Access it now though the Monitor menu in the Power Automate portal:

This new monitoring interface allows makers and admins to track the data processed by AI Builder models, as well as the processing time and number of credits consumed by flow or app runs. Administrators of the environment can access data for all users and models in the environment, while makers will be able to access monitoring activity generated by their own usage.

The data is stored in the AI Event Dataverse table, giving flexibility to generate custom reports using Power BI so that you can tailor monitoring views for your specific need.

Learn more about this new feature in the Activity monitoring (preview) | Microsoft Learn page.

Extend Invoice Processing with custom fields

You can now extend the behavior of the prebuilt invoice processing model by adding new fields to be extracted in addition to the default ones. Start by creating a custom document processing model, then select “Invoices (preview)” as the type of document to process.

Instead of having to build an invoice processing model from scratch, you can re-use all the default fields of the invoice processing pre-built model and add only the additional ones you need. This feature will simplify the model creation process for custom invoices, making it much easier to train a custom model.

Learn more about this feature: Create a document processing custom model in AI Builder | Microsoft Learn.

Co-ownership of AI models

AI Builder now supports multiple users co-owning an AI model, empowering multiple individuals to contribute to the AI model training lifecycle process. By enabling multiple owners, model improvements can be performed by multiple individuals within a company, allowing teams of makers to update and improve model performance over time.

In the model detail page of a model, users now have a new option to share as “Co-owner”. Selecting it when you want to share the model, will grant edit rights to the users that you define:

Learn more about this new feature: Share your AI model – AI Builder | Microsoft Learn.

Pricing and more

AI Builder prices have also been adjusted, with updated rates and volume discounts available. Download the Power Platform Licensing Guide to see the latest updates to AI Builder pricing.

Please feel free to provide your AI Builder questions and feedback in the Power Automate Community. Happy Automation!

The post July 2023 updates for AI Builder appeared first on Microsoft Power Platform Blog.

In the following 1-minute video see how you can use the new GPT action in AI Builder to extract the data your need from the body of the email and save the results in a CSV file. No advanced technical knowledge is required!

The new GPT AI model in AI Builder lets you do amazing things like this and more.

We’re always excited to see your feedback, so drop us a comment on the video. Happy automation!

The post Automate email data extraction in minutes with Power Automate, AI Builder, and GPT appeared first on Microsoft Power Platform Blog.