{"id":1174801,"date":"2026-06-09T01:52:53","date_gmt":"2026-06-09T08:52:53","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-event&#038;p=1174801"},"modified":"2026-06-09T01:57:40","modified_gmt":"2026-06-09T08:57:40","slug":"global-ai-value-challenge","status":"publish","type":"msr-event","link":"https:\/\/www.microsoft.com\/en-us\/research\/event\/global-ai-value-challenge\/","title":{"rendered":"Global AI Value Challenge"},"content":{"rendered":"\n\n\n\n\n<p id=\"global-ai-values-challengeadvancing-research-on-ai-aligned-with-human-values\">AI systems are increasingly used in science, technology, and everyday decision-making. But a harder problem is often overlooked: when an honesty conflicts with kindness, how should an AI system respond? In situations with limited or ambiguous information, can these systems make decisions that reflect human values and ethical judgment?<\/p>\n\n\n\n<p>To study this problem, Microsoft Research Asia is launching the <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">Global AI Values Challenge<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>, an open research initiative focused on value alignment in realistic scenarios. <a>The challenge is open for submission through August 31, 2026, <\/a>and invites researchers to evaluate whether AI systems can reason, judge, and act in ways that align with human values when faced with ethical dilemmas.<\/p>\n\n\n<aside id=accordion-abb81179-527d-416b-8729-daf791534609 class=\"msr-table-of-contents-block accordion mb-5 pb-0\" data-bi-aN=\"table-of-contents\">\n\t<button class=\"btn btn-collapse bg-gray-100 mb-0 display-flex justify-content-between\" type=\"button\" data-mount=\"collapse\" data-target=\"#accordion-collapse-abb81179-527d-416b-8729-daf791534609\" aria-expanded=\"true\" aria-controls=\"accordion-collapse-abb81179-527d-416b-8729-daf791534609\">\n\t\t<span class=\"msr-table-of-contents-block__label subtitle\">In this article<\/span>\n\t\t<span class=\"msr-table-of-contents-block__current mr-4 text-gray-600 font-weight-normal\" aria-hidden=\"true\"><\/span>\n\t<\/button>\n\t<div id=\"accordion-collapse-abb81179-527d-416b-8729-daf791534609\" class=\"msr-table-of-contents-block__collapse-wrapper collapse show\" data-parent=\"#accordion-abb81179-527d-416b-8729-daf791534609\">\n\t\t<div class=\"accordion-body bg-gray-100 border-top pt-4\">\n\t\t\t<ol class=\"msr-table-of-contents-block__list\">\n\t\t\t\t\t\t\t\t\t<li class=\"msr-table-of-contents-block__list-item\">\n\t\t\t\t\t\t<a href=\"#advancing-research-on-ai-aligned-with-human-values\" class=\"msr-table-of-contents-block__list-item-link\">Advancing research on AI aligned with human values<\/a>\n\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t\t<li class=\"msr-table-of-contents-block__list-item\">\n\t\t\t\t\t\t<a href=\"#who-should-participate\" class=\"msr-table-of-contents-block__list-item-link\">Who should participate<\/a>\n\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t\t<li class=\"msr-table-of-contents-block__list-item\">\n\t\t\t\t\t\t<a href=\"#how-to-participate-1\" class=\"msr-table-of-contents-block__list-item-link\">How to participate<\/a>\n\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t\t<li class=\"msr-table-of-contents-block__list-item\">\n\t\t\t\t\t\t<a href=\"#how-submissions-are-evaluated\" class=\"msr-table-of-contents-block__list-item-link\">How submissions are evaluated<\/a>\n\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t\t<li class=\"msr-table-of-contents-block__list-item\">\n\t\t\t\t\t\t<a href=\"#benefits-of-participation\" class=\"msr-table-of-contents-block__list-item-link\">Benefits of participation<\/a>\n\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t\t<li class=\"msr-table-of-contents-block__list-item\">\n\t\t\t\t\t\t<a href=\"#research-context-societal-ai\" class=\"msr-table-of-contents-block__list-item-link\">Research context: Societal AI<\/a>\n\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t\t\t<li class=\"msr-table-of-contents-block__list-item\">\n\t\t\t\t\t\t<a href=\"#defining-the-future-of-ai\" class=\"msr-table-of-contents-block__list-item-link\">Defining the future of AI<\/a>\n\t\t\t\t\t<\/li>\n\t\t\t\t\t\t\t<\/ul>\n\t\t<\/div>\n\t<\/div>\n\t<span class=\"msr-table-of-contents-block__progress-bar\"><\/span>\n<\/aside>\n\n\n\n<h2 class=\"wp-block-heading is-style-l\" id=\"advancing-research-on-ai-aligned-with-human-values\"><strong>Advancing research on AI aligned with human values<\/strong><\/h2>\n\n\n\n<p>Rather than focusing on benchmarks such as coding performance or mathematical reasoning, the challenge centers on value-based reasoning. Participants are invited to propose realistic dilemmas that reveal where current AI systems struggle to make consistent or well-justified decisions.<\/p>\n\n\n\n<p>Submissions must be grounded in real-world scenarios, not abstract thought experiments. The goal is to build a high-quality global dataset of questions and answers about ethical decision-making to support research in AI alignment with moral reasoning.<\/p>\n\n\n\n<h3 class=\"wp-block-heading is-style-m\" id=\"who-should-participate\"><strong>Who should participate<\/strong><\/h3>\n\n\n\n<p>The challenge is open to the global academic and research community, including:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Students, postdoctoral researchers, and faculty members<\/li>\n\n\n\n<li>Researchers and practitioners in fields such as psychology, sociology, philosophy, ethics, law, literature, journalism, communications, and computational social science<\/li>\n\n\n\n<li>Individuals working at the intersection of AI and the social sciences<\/li>\n<\/ul>\n\n\n\n<p>Understanding human values in complex scenarios requires input across cultures and disciplines. The challenge, therefore, invites participation from a broad range of fields.<\/p>\n\n\n\n<h3 class=\"wp-block-heading is-style-m\" id=\"how-to-participate-1\"><strong>How to participate<\/strong><\/h3>\n\n\n\n<ol start=\"1\" class=\"wp-block-list\">\n<li><strong>Register:<\/strong> Between May 7 and August 31, 2026, visit the official <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">website<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> and complete registration using an organizational email address. Both individuals and teams are welcome to register.<\/li>\n\n\n\n<li><strong>Submit:<\/strong> Contribute original human values or ethics related dilemmas, along with proposed answers and supporting justification. Submissions should be provided as natural-language Q&A pairs. Additional guidelines and submission requirements are available in the event <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">website<span class=\"sr-only\"> (opens in new tab)<\/span><\/a> after registration.<\/li>\n\n\n\n<li><strong>Recognition:<\/strong> Selected submissions will be included in the challenge dataset and may receive recognition.<\/li>\n<\/ol>\n\n\n\n<h3 class=\"wp-block-heading is-style-m\" id=\"how-submissions-are-evaluated\"><strong>How submissions are evaluated<\/strong><\/h3>\n\n\n\n<p>Submissions will be evaluated through a two-stage review process designed to assess both the difficulty of the dilemma for current AI systems and the quality of the ethical reasoning.<\/p>\n\n\n\n<p>In the first stage, submissions are reviewed using AI models to determine whether the dilemma presents a meaningful challenge. Only submissions that models cannot answer correctly advance to the second stage.<\/p>\n\n\n\n<p>In the second stage, domain experts, from academia in fields such as psychology, philosophy, and computer science, evaluate submissions based on criteria such as difficulty, originality, logical soundness, and the quality of the supporting justification. Final decisions are made at this stage.<\/p>\n\n\n\n<p>Accepted submissions may be included in the challenge dataset to support future research on AI value alignment and ethical reasoning. Participants will be notified of the outcome through their registered email address.<\/p>\n\n\n\n<h3 class=\"wp-block-heading is-style-m\" id=\"benefits-of-participation\"><strong>Benefits of participation<\/strong><\/h3>\n\n\n\n<p>Participants may receive:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li>Monetary awards of RMB 200 or RMB 500 for accepted submissions<\/li>\n\n\n\n<li>A certificate of participation<\/li>\n\n\n\n<li>Recognition as a contributor to the related dataset<\/li>\n\n\n\n<li>Opportunities for internships or research collaboration with the MSRA Societal AI team<\/li>\n<\/ul>\n\n\n\n<p>**Final rules, eligibility requirements, and award details are subject to the information published on <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">the official challenge website<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading is-style-l\" id=\"research-context-societal-ai\"><strong>Research context: Societal AI<\/strong><\/h2>\n\n\n\n<p>As large language models (LLMs) become more integrated into real-world systems, they are increasingly involved in decisions with ethical and societal implications. These systems must navigate ambiguity, trade-offs, and context-dependent judgments.<\/p>\n\n\n\n<p>Microsoft Research Asia\u2019s Societal AI group examines how AI systems can move beyond producing correct answers to making decisions that are reasonable and fair in context. Prior work, including the \u201c<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2025\/03\/Societal-AI-Research-Challenges-and-Opportunities.pdf\">Societal AI: Research Challenges and Opportunities<\/a>\u201d white paper and the <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/value-compass\/\">Value Compass<\/a> project, explores ways to represent and evaluate human values in computational systems.<\/p>\n\n\n\n<p>\u201cAI development has reached a critical turning point. We must not only pursue the upper limits of AI capability but also uphold its ethical foundation,\u201d said Lidong Zhou, corporate vice president at Microsoft and managing director of Microsoft Research Asia.<\/p>\n\n\n\n<p>\u201cValue alignment is not only a technical challenge but also a social one,\u201d said Xing Xie, assistant managing director of Microsoft Research Asia. \u201cThrough efforts like the Value Compass project and the Global AI Values Challenge, we aim to incorporate insights from the social sciences into AI systems.\u201d<\/p>\n\n\n\n<h2 class=\"wp-block-heading is-style-l\" id=\"defining-the-future-of-ai\"><strong>Defining the future of AI<\/strong><\/h2>\n\n\n\n<p>From recommendation systems in news media to clinical decision support in healthcare, and from financial risk assessment to autonomous driving, AI systems are increasingly embedded in real-world settings where their decisions affect individuals and society.<\/p>\n\n\n\n<p>The Global AI Values Challenge brings together perspectives from across disciplines to better understand how AI systems respond in complex situations. By contributing realistic dilemmas and well-reasoned responses, participants can help advance research on AI alignment with human values.<\/p>\n\n\n\n<p>Registration is now open on the official <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">website<span class=\"sr-only\"> (opens in new tab)<\/span><\/a>.<\/p>\n\n\n\n\n\n<p>Register in the <a class=\"msr-external-link glyph-append glyph-append-open-in-new-tab glyph-append-xsmall\" rel=\"noopener noreferrer\" target=\"_blank\" href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">Global AI Values Challenge<span class=\"sr-only\"> (opens in new tab)<\/span><\/a><\/p>\n\n\n\n<div class=\"wp-block-media-text has-vertical-margin-small  has-vertical-padding-none  is-stacked-on-mobile\" style=\"grid-template-columns:37% auto\"><figure class=\"wp-block-media-text__media\"><img loading=\"lazy\" decoding=\"async\" width=\"686\" height=\"1024\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/image-3-686x1024.png\" alt=\"diagram\" class=\"wp-image-1175125 size-full\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/image-3-686x1024.png 686w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/image-3-201x300.png 201w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/image-3-768x1146.png 768w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/image-3-121x180.png 121w\" sizes=\"auto, (max-width: 686px) 100vw, 686px\" \/><\/figure><div class=\"wp-block-media-text__content\">\n<p><\/p>\n<\/div><\/div>\n\n\n\n<p>Pls visit our official website to register and learn more about the Challenge.<\/p>\n\n\n","protected":false},"excerpt":{"rendered":"<p>AI systems are increasingly used in science, technology, and everyday decision-making. But a harder problem is often overlooked: when an honesty conflicts with kindness, how should an AI system respond? In situations with limited or ambiguous information, can these systems make decisions that reflect human values and ethical judgment? To study this problem, Microsoft Research [&hellip;]<\/p>\n","protected":false},"featured_media":1174804,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_startdate":"2026-05-06","msr_enddate":"2026-08-31","msr_location":"Global online","msr_expirationdate":"","msr_event_recording_link":"","msr_event_link":"","msr_event_link_redirect":false,"msr_event_time":"","msr_hide_region":false,"msr_private_event":false,"msr_hide_image_in_river":0,"footnotes":""},"research-area":[13556,13554,13553,13559],"msr-region":[],"msr-event-type":[197944],"msr-video-type":[],"msr-locale":[268875],"msr-program-audience":[],"msr-post-option":[],"msr-impact-theme":[],"class_list":["post-1174801","msr-event","type-msr-event","status-publish","has-post-thumbnail","hentry","msr-research-area-artificial-intelligence","msr-research-area-human-computer-interaction","msr-research-area-medical-health-genomics","msr-research-area-social-sciences","msr-event-type-hosted-by-microsoft","msr-locale-en_us"],"msr_about":"<!-- wp:msr\/event-details {\"title\":\"Global AI Value Challenge\",\"subTitle\":\"\\u003cins\\u003eAdvancing research on \\u003c\/ins\\u003e \\u003cins\\u003eAI aligned with human values\\u003c\/ins\\u003e\",\"image\":{\"id\":1174804,\"url\":\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/Weixin-Image_20260527160725_10_7534-1-scaled-e1780645348602.jpg\",\"alt\":\"chart\"}} \/-->\n\n<!-- wp:msr\/content-tabs -->\n<!-- wp:msr\/content-tab -->\n<!-- wp:paragraph -->\n<p id=\"global-ai-values-challengeadvancing-research-on-ai-aligned-with-human-values\">AI systems are increasingly used in science, technology, and everyday decision-making. But a harder problem is often overlooked: when an honesty conflicts with kindness, how should an AI system respond? In situations with limited or ambiguous information, can these systems make decisions that reflect human values and ethical judgment?<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>To study this problem, Microsoft Research Asia is launching the <a href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">Global AI Values Challenge<\/a>, an open research initiative focused on value alignment in realistic scenarios. <a>The challenge is open for submission through August 31, 2026, <\/a>and invites researchers to evaluate whether AI systems can reason, judge, and act in ways that align with human values when faced with ethical dilemmas.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:msr\/block-table-of-contents {\"id\":\"abb81179-527d-416b-8729-daf791534609\",\"levelsAllowed\":[2,3,4,5,6],\"headers\":[{\"level\":2,\"anchor\":\"#advancing-research-on-ai-aligned-with-human-values\",\"content\":\"Advancing research on AI aligned with human values\"},{\"level\":3,\"anchor\":\"#who-should-participate\",\"content\":\"Who should participate\"},{\"level\":3,\"anchor\":\"#how-to-participate-1\",\"content\":\"How to participate\"},{\"level\":3,\"anchor\":\"#how-submissions-are-evaluated\",\"content\":\"How submissions are evaluated\"},{\"level\":3,\"anchor\":\"#benefits-of-participation\",\"content\":\"Benefits of participation\"},{\"level\":2,\"anchor\":\"#research-context-societal-ai\",\"content\":\"Research context: Societal AI\"},{\"level\":2,\"anchor\":\"#defining-the-future-of-ai\",\"content\":\"Defining the future of AI\"}]} \/-->\n\n<!-- wp:heading {\"className\":\"is-style-l\"} -->\n<h2 class=\"wp-block-heading is-style-l\" id=\"advancing-research-on-ai-aligned-with-human-values\"><strong>Advancing research on AI aligned with human values<\/strong><\/h2>\n<!-- \/wp:heading -->\n\n<!-- wp:paragraph -->\n<p>Rather than focusing on benchmarks such as coding performance or mathematical reasoning, the challenge centers on value-based reasoning. Participants are invited to propose realistic dilemmas that reveal where current AI systems struggle to make consistent or well-justified decisions.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>Submissions must be grounded in real-world scenarios, not abstract thought experiments. The goal is to build a high-quality global dataset of questions and answers about ethical decision-making to support research in AI alignment with moral reasoning.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:heading {\"level\":3,\"className\":\"is-style-m\"} -->\n<h3 class=\"wp-block-heading is-style-m\" id=\"who-should-participate\"><strong>Who should participate<\/strong><\/h3>\n<!-- \/wp:heading -->\n\n<!-- wp:paragraph -->\n<p>The challenge is open to the global academic and research community, including:<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:list -->\n<ul class=\"wp-block-list\"><!-- wp:list-item -->\n<li>Students, postdoctoral researchers, and faculty members<\/li>\n<!-- \/wp:list-item -->\n\n<!-- wp:list-item -->\n<li>Researchers and practitioners in fields such as psychology, sociology, philosophy, ethics, law, literature, journalism, communications, and computational social science<\/li>\n<!-- \/wp:list-item -->\n\n<!-- wp:list-item -->\n<li>Individuals working at the intersection of AI and the social sciences<\/li>\n<!-- \/wp:list-item --><\/ul>\n<!-- \/wp:list -->\n\n<!-- wp:paragraph -->\n<p>Understanding human values in complex scenarios requires input across cultures and disciplines. The challenge, therefore, invites participation from a broad range of fields.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:heading {\"level\":3,\"className\":\"is-style-m\"} -->\n<h3 class=\"wp-block-heading is-style-m\" id=\"how-to-participate-1\"><strong>How to participate<\/strong><\/h3>\n<!-- \/wp:heading -->\n\n<!-- wp:list {\"ordered\":true,\"start\":1} -->\n<ol start=\"1\" class=\"wp-block-list\"><!-- wp:list-item -->\n<li><strong>Register:<\/strong> Between May 7 and August 31, 2026, visit the official <a href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">website<\/a> and complete registration using an organizational email address. Both individuals and teams are welcome to register.<\/li>\n<!-- \/wp:list-item -->\n\n<!-- wp:list-item -->\n<li><strong>Submit:<\/strong> Contribute original human values or ethics related dilemmas, along with proposed answers and supporting justification. Submissions should be provided as natural-language Q&amp;A pairs. Additional guidelines and submission requirements are available in the event <a href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">website<\/a> after registration.<\/li>\n<!-- \/wp:list-item -->\n\n<!-- wp:list-item -->\n<li><strong>Recognition:<\/strong> Selected submissions will be included in the challenge dataset and may receive recognition.<\/li>\n<!-- \/wp:list-item --><\/ol>\n<!-- \/wp:list -->\n\n<!-- wp:heading {\"level\":3,\"className\":\"is-style-m\"} -->\n<h3 class=\"wp-block-heading is-style-m\" id=\"how-submissions-are-evaluated\"><strong>How submissions are evaluated<\/strong><\/h3>\n<!-- \/wp:heading -->\n\n<!-- wp:paragraph -->\n<p>Submissions will be evaluated through a two-stage review process designed to assess both the difficulty of the dilemma for current AI systems and the quality of the ethical reasoning.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>In the first stage, submissions are reviewed using AI models to determine whether the dilemma presents a meaningful challenge. Only submissions that models cannot answer correctly advance to the second stage.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>In the second stage, domain experts, from academia in fields such as psychology, philosophy, and computer science, evaluate submissions based on criteria such as difficulty, originality, logical soundness, and the quality of the supporting justification. Final decisions are made at this stage.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>Accepted submissions may be included in the challenge dataset to support future research on AI value alignment and ethical reasoning. Participants will be notified of the outcome through their registered email address.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:heading {\"level\":3,\"className\":\"is-style-m\"} -->\n<h3 class=\"wp-block-heading is-style-m\" id=\"benefits-of-participation\"><strong>Benefits of participation<\/strong><\/h3>\n<!-- \/wp:heading -->\n\n<!-- wp:paragraph -->\n<p>Participants may receive:<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:list -->\n<ul class=\"wp-block-list\"><!-- wp:list-item -->\n<li>Monetary awards of RMB 200 or RMB 500 for accepted submissions<\/li>\n<!-- \/wp:list-item -->\n\n<!-- wp:list-item -->\n<li>A certificate of participation<\/li>\n<!-- \/wp:list-item -->\n\n<!-- wp:list-item -->\n<li>Recognition as a contributor to the related dataset<\/li>\n<!-- \/wp:list-item -->\n\n<!-- wp:list-item -->\n<li>Opportunities for internships or research collaboration with the MSRA Societal AI team<\/li>\n<!-- \/wp:list-item --><\/ul>\n<!-- \/wp:list -->\n\n<!-- wp:paragraph -->\n<p>**Final rules, eligibility requirements, and award details are subject to the information published on <a href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">the official challenge website<\/a>.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:heading {\"className\":\"is-style-l\"} -->\n<h2 class=\"wp-block-heading is-style-l\" id=\"research-context-societal-ai\"><strong>Research context: Societal AI<\/strong><\/h2>\n<!-- \/wp:heading -->\n\n<!-- wp:paragraph -->\n<p>As large language models (LLMs) become more integrated into real-world systems, they are increasingly involved in decisions with ethical and societal implications. These systems must navigate ambiguity, trade-offs, and context-dependent judgments.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>Microsoft Research Asia\u2019s Societal AI group examines how AI systems can move beyond producing correct answers to making decisions that are reasonable and fair in context. Prior work, including the \u201c<a href=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2025\/03\/Societal-AI-Research-Challenges-and-Opportunities.pdf\">Societal AI: Research Challenges and Opportunities<\/a>\u201d white paper and the <a href=\"https:\/\/www.microsoft.com\/en-us\/research\/project\/value-compass\/\">Value Compass<\/a> project, explores ways to represent and evaluate human values in computational systems.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>\u201cAI development has reached a critical turning point. We must not only pursue the upper limits of AI capability but also uphold its ethical foundation,\u201d said Lidong Zhou, corporate vice president at Microsoft and managing director of Microsoft Research Asia.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>\u201cValue alignment is not only a technical challenge but also a social one,\u201d said Xing Xie, assistant managing director of Microsoft Research Asia. \u201cThrough efforts like the Value Compass project and the Global AI Values Challenge, we aim to incorporate insights from the social sciences into AI systems.\u201d<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:heading {\"className\":\"is-style-l\"} -->\n<h2 class=\"wp-block-heading is-style-l\" id=\"defining-the-future-of-ai\"><strong>Defining the future of AI<\/strong><\/h2>\n<!-- \/wp:heading -->\n\n<!-- wp:paragraph -->\n<p>From recommendation systems in news media to clinical decision support in healthcare, and from financial risk assessment to autonomous driving, AI systems are increasingly embedded in real-world settings where their decisions affect individuals and society.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>The Global AI Values Challenge brings together perspectives from across disciplines to better understand how AI systems respond in complex situations. By contributing realistic dilemmas and well-reasoned responses, participants can help advance research on AI alignment with human values.<\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:paragraph -->\n<p>Registration is now open on the official <a href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">website<\/a>.<\/p>\n<!-- \/wp:paragraph -->\n<!-- \/wp:msr\/content-tab -->\n\n<!-- wp:msr\/content-tab {\"title\":\"Register Now\"} -->\n<!-- wp:paragraph -->\n<p>Register in the <a href=\"https:\/\/microsoft.github.io\/ValueChallenge\/\">Global AI Values Challenge<\/a><\/p>\n<!-- \/wp:paragraph -->\n\n<!-- wp:media-text {\"mediaId\":1175125,\"mediaLink\":\"https:\/\/www.microsoft.com\/en-us\/research\/?attachment_id=1175125\",\"mediaType\":\"image\",\"mediaWidth\":37,\"backgroundColor\":\"\"} -->\n<div class=\"wp-block-media-text is-stacked-on-mobile\" style=\"grid-template-columns:37% auto\"><figure class=\"wp-block-media-text__media\"><img src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/image-3-686x1024.png\" alt=\"diagram\" class=\"wp-image-1175125 size-full\"\/><\/figure><div class=\"wp-block-media-text__content\"><!-- wp:paragraph {\"placeholder\":\"Content\u2026\"} -->\n<p><\/p>\n<!-- \/wp:paragraph --><\/div><\/div>\n<!-- \/wp:media-text -->\n\n<!-- wp:paragraph -->\n<p>Pls visit our official website to register and learn more about the Challenge.<\/p>\n<!-- \/wp:paragraph -->\n<!-- \/wp:msr\/content-tab -->\n<!-- \/wp:msr\/content-tabs -->","tab-content":[],"msr_startdate":"2026-05-06","msr_enddate":"2026-08-31","msr_event_time":"","msr_location":"Global online","msr_event_link":"","msr_event_recording_link":"","msr_startdate_formatted":"May 6, 2026","msr_register_text":"Watch now","msr_cta_link":"","msr_cta_text":"","msr_cta_bi_name":"","featured_image_thumbnail":"<img width=\"960\" height=\"540\" src=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/Weixin-Image_20260527160725_10_7534-1-scaled-e1780645348602-960x540.jpg\" class=\"img-object-cover\" alt=\"chart\" decoding=\"async\" loading=\"lazy\" srcset=\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/Weixin-Image_20260527160725_10_7534-1-scaled-e1780645348602-960x540.jpg 960w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/Weixin-Image_20260527160725_10_7534-1-scaled-e1780645348602-1066x600.jpg 1066w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/Weixin-Image_20260527160725_10_7534-1-scaled-e1780645348602-655x368.jpg 655w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/Weixin-Image_20260527160725_10_7534-1-scaled-e1780645348602-640x360.jpg 640w, https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2026\/06\/Weixin-Image_20260527160725_10_7534-1-scaled-e1780645348602-1280x720.jpg 1280w\" sizes=\"auto, (max-width: 960px) 100vw, 960px\" \/>","event_excerpt":"AI systems are increasingly used in science, technology, and everyday decision-making. But a harder problem is often overlooked: when an honesty conflicts with kindness, how should an AI system respond? In situations with limited or ambiguous information, can these systems make decisions that reflect human values and ethical judgment? To study this problem, Microsoft Research Asia is launching the Global AI Values Challenge (opens in new tab), an open research initiative focused on value alignment&hellip;","msr_research_lab":[199560],"related-researchers":[],"msr_impact_theme":[],"related-academic-programs":[],"related-groups":[],"related-projects":[995412],"related-opportunities":[],"related-publications":[],"related-videos":[],"related-posts":[],"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/1174801","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-event"}],"version-history":[{"count":20,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/1174801\/revisions"}],"predecessor-version":[{"id":1175131,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event\/1174801\/revisions\/1175131"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/1174804"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=1174801"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=1174801"},{"taxonomy":"msr-region","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-region?post=1174801"},{"taxonomy":"msr-event-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-event-type?post=1174801"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=1174801"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=1174801"},{"taxonomy":"msr-program-audience","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-program-audience?post=1174801"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=1174801"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=1174801"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}