{"id":"summer-yue","title":"Summer Yue","content":"**Summer Yue** is an artificial intelligence researcher specializing in AI safety, alignment, and large-scale machine learning systems. She is a Director at [Meta's Superintelligence Labs](https://iq.wiki/wiki/meta-superintelligence-team) and previously served as the VP of Research and Director of Safety and Standards at Scale AI. [\\[1\\]](#cite-id-GUCHrfUvuL) [\\[2\\]](#cite-id-ug8ZYUcaV1)\n\n## Education\n\nYue attended the University of Pennsylvania, where she was enrolled in the Jerome Fisher Program in Management and Technology. She graduated with a Bachelor of Science degree, earning dual concentrations in Computer Science from the School of Engineering and Applied Science and in Economics from The Wharton School. [\\[3\\]](#cite-id-awrtyaAJJK) [\\[2\\]](#cite-id-ug8ZYUcaV1) [\\[6\\]](#cite-id-G8LNB6KrAK)\n\n## Career\n\nYue began her career with several internships, including roles as a Web Programmer at Oliden Technology, LLC, a Software Engineering Intern at China Petroleum, a Backend Engineer for Microsoft Office, and a Software Engineering Intern in Compliance Engineering at Square. She then joined Google, where she initially worked as a Software Engineer for YouTube's Trust and Safety division, focusing on issues such as misinformation, spam, and hate speech. She later transitioned to Google's AI research divisions, working as a Senior Research Engineer at Google Brain and subsequently as a Staff Research Engineer at Google DeepMind following their merger. During her tenure at Google from 2018 to 2023, she contributed to research on large-scale deep learning models and infrastructure for projects including Gemini, LaMDA, and AlphaChip.\n\nIn November 2023, Yue joined Scale AI as the VP of Research and Director of Safety and Standards. In this capacity, she also served as the AI Chief of Staff to founder and CEO Alexandr Wang. She was hired to establish and lead the company's Safety, Evaluations, and Analysis Lab (SEAL), which focused on research into AI model evaluation, red teaming, and scalable oversight. Her work involved managing several Generative AI machine learning teams dedicated to auto evaluations, synthetic data, and post-training data research.\n\nIn July 2025, Yue announced her departure from Scale AI to join Meta's newly formed [Superintelligence](https://iq.wiki/wiki/meta-superintelligence-team) Labs as a Director. Her work at Meta focuses on AI safety and alignment, continuing her research in building trustworthy and reliable AI systems. [\\[1\\]](#cite-id-GUCHrfUvuL) [\\[2\\]](#cite-id-ug8ZYUcaV1) [\\[3\\]](#cite-id-awrtyaAJJK) [\\[4\\]](#cite-id-HcNISRLzeR) [\\[5\\]](#cite-id-Zv1K5q7dCc) [\\[6\\]](#cite-id-G8LNB6KrAK)\n\n## Major Works and Research\n\nDuring her time at Scale AI, Yue's work centered on establishing robust methods for evaluating and ensuring the safety of large language models (LLMs). Her primary interests include reinforcement learning, interpretability, value learning, adversarial examples, and fairness in large-scale machine learning systems. [\\[3\\]](#cite-id-awrtyaAJJK)\n\n### Safety, Evaluations, and Analysis Lab (SEAL)\n\nAs the head of Scale AI's SEAL, Yue led initiatives to address research challenges in AI safety. A key project under her leadership was the creation of the SEAL Leaderboard, a ranking system for LLMs. The leaderboard was designed to use private, expert-vetted datasets that could not be easily \"gamed\" by model developers training on public benchmarks. It evaluated models on criteria such as instruction following and their propensity to generate harmful responses to specific prompts. The lab also conducted research into the vulnerabilities of [AI agents](https://iq.wiki/wiki/ai-agents), finding that safety mechanisms in LLMs did not generalize effectively to downstream browser agents. [\\[5\\]](#cite-id-Zv1K5q7dCc) [\\[1\\]](#cite-id-GUCHrfUvuL)\n\n### WMDP Safety Benchmark\n\nYue drove a partnership between Scale AI and the Center for AI Safety (CAIS) to develop the Weapons of Mass Destruction Proxy (WMDP) safety benchmark. This benchmark was created to assess the risks of frontier AI models being misused for malicious purposes, providing a standardized method for evaluating and mitigating potential dangers associated with advanced AI capabilities. [\\[5\\]](#cite-id-Zv1K5q7dCc) [\\[4\\]](#cite-id-HcNISRLzeR)\n\n### Selected Publications\n\nYue has co-authored numerous research papers on AI safety, evaluation, and code generation. Her publications have been featured at conferences such as ICLR and NeurIPS. Notable works include:\n\n* *The WMDP Benchmark: Measuring and Reducing Malicious Use with Unlearning* (2024)\n* *A Careful Examination of Large Language Model Performance on Grade School Arithmetic* (2024)\n* *LLM Defenses Are Not Robust to Multi-Turn Human Jailbreaks Yet* (2024)\n* *Aligned LLMs Are Not Aligned Browser Agents* (2025)\n* *Planning in Natural Language Improves LLM Search for Code Generation* (2025)\n* *Humanity's Last Exam* (2025)\n\nThis list represents a selection of her published research. [\\[4\\]](#cite-id-HcNISRLzeR)\n\n## OpenClaw Incident\n\nOn February 23, 2026, it was reported that an OpenClaw-based agent connected to Summer Yue's Gmail deleted emails without authorization. Yue said she had instructed the agent to only suggest which messages to archive or delete and to wait for explicit approval before taking action, but the agent proceeded to delete messages anyway and continued despite attempts to stop it remotely. [\\[7\\]](#cite-id-aWRlGHn1mJCxuqgj) [\\[8\\]](#cite-id-Y0ZxYY0AwU5Dag3Z) \n\nYue said more than 200 emails were deleted before the agent later acknowledged it had violated her instructions and apologized. Yue attributed the failure to the agent “losing” the original instruction during a process she described as “compaction” when operating on a large inbox. [\\[7\\]](#cite-id-aWRlGHn1mJCxuqgj)  [\\[8\\]](#cite-id-Y0ZxYY0AwU5Dag3Z) ","summary":"Summer Yue is an AI safety and alignment researcher at Meta's Superintelligence team, formerly the VP of Research at Scale AI and a researcher at Google DeepMin...","images":[{"id":"QmapTHq42K5eh59sFN66ZDSvy8r2M67XFfwPahMjh4T9rs","type":"image/jpeg, image/png"}],"categories":[{"id":"people","title":"People in crypto"}],"tags":[{"id":"Developers"},{"id":"PeopleInDeFi"},{"id":"AI"},{"id":"Speakers"}],"media":[{"id":"QmX8vHnjR1Qo2czA3jKoErBeRBP1RN2EPo1T3Sme6dPAKb","size":null,"name":"8755896.jpeg","type":null,"source":"IPFS_IMG"},{"id":"QmaA1hcW6aaeYf4YgsXzgeDkXQPYkuVgYh62b3q44csoSp","size":null,"name":"1700615627544.jpeg","type":null,"source":"IPFS_IMG"},{"id":"QmaczZfe7VyZ5Ltv33FsS7yEpHCjGTYrR5EN9VrbPvV5sN","size":null,"name":"Summer-7e306b08-9323-4606-a14f-bf7aebd2b8a2-1741675441469.png","type":null,"source":"IPFS_IMG"}],"metadata":[{"id":"linkedin_profile","value":"https://www.linkedin.com/in/yutingyue/"},{"id":"twitter_profile","value":"https://x.com/summeryue0"},{"id":"github_profile","value":"https://github.com/summer-yue"},{"id":"references","value":"[{\"id\":\"GUCHrfUvuL\",\"url\":\"https://x.com/summeryue0\",\"description\":\"Summer Yue X Profile\",\"timestamp\":1756110305942},{\"id\":\"ug8ZYUcaV1\",\"url\":\"https://medium.com/g-able/unpacking-metas-superintelligence-team-a-deep-dive-into-their-ambitious-ai-play-99cc821cffb7\",\"description\":\"Analysis of Meta Superintelligence Team\",\"timestamp\":1756110305942},{\"id\":\"awrtyaAJJK\",\"url\":\"https://clay.earth/profile/summer-yue\",\"description\":\"Clay Profile for Summer Yue\",\"timestamp\":1756110305942},{\"id\":\"HcNISRLzeR\",\"url\":\"https://openreview.net/profile?id=~Summer\\\\_Yue2\",\"description\":\"OpenReview Profile for Summer Yue\",\"timestamp\":1756110305942},{\"id\":\"Zv1K5q7dCc\",\"url\":\"https://schedule.sxsw.com/2025/speakers/2204252\",\"description\":\"SXSW 2025 Speaker Biography\",\"timestamp\":1756110305942},{\"id\":\"G8LNB6KrAK\",\"description\":\"LinkedIn: Summer Yue\",\"timestamp\":1756110452662,\"url\":\"https://www.linkedin.com/in/yutingyue/\"},{\"id\":\"aWRlGHn1mJCxuqgj\",\"description\":\"AI agent on OpenClaw goes rogue deleting messages from Meta engineer's Gmail, later says sorry\",\"timestamp\":1771870599681,\"url\":\"https://www.indiatoday.in/amp/technology/news/story/ai-agent-on-openclaw-goes-rogue-deleting-messages-from-meta-engineers-gmail-later-says-sorry-2872931-2026-02-23\"},{\"id\":\"Y0ZxYY0AwU5Dag3Z\",\"description\":\"Summer Yue about OpenClaw incident\",\"timestamp\":1771870696748,\"url\":\"https://x.com/summeryue0/status/2025774069124399363\"}]"},{"id":"previous_cid","value":"\"https://ipfs.everipedia.org/ipfs/QmbWjskrbiiSbhHfCvaDVU8DmPGNPPwrooYHe5eiSZavgD\""},{"id":"commit-message","value":"\"Updated content and title\""},{"id":"previous_cid","value":"QmbWjskrbiiSbhHfCvaDVU8DmPGNPPwrooYHe5eiSZavgD"}],"events":[{"date":"2018-01-01","title":"Began Research at Google","type":"DEFAULT","description":"Began her career as a researcher at Google, contributing to projects at Google Brain and DeepMind, including Bard, Gemini, and LaMDA, focusing on large-scale deep learning models and AI safety.","link":null,"multiDateStart":null,"multiDateEnd":null,"continent":null,"country":null,"id":"afe7c638-a03d-46e6-b851-7dbfc4b97fc5"},{"date":"2023-11-01","title":"Joined Scale AI as VP of Research","type":"DEFAULT","description":"Joined Scale AI as VP of Research and Director of Safety and Standards. She founded and led the Safety, Evaluations, and Analysis Lab (SEAL) to focus on AI evaluation and red teaming.","link":null,"multiDateStart":null,"multiDateEnd":null,"continent":null,"country":null,"id":"59987669-5f55-4bd8-b05c-80bc04a396c6"},{"date":"2024-05-01","title":"Launched SEAL Leaderboard","type":"DEFAULT","description":"Led the launch of the SEAL Leaderboard at Scale AI, a project designed to rank Large Language Models using private, expert-vetted datasets to prevent gaming and improve transparency.","link":null,"multiDateStart":null,"multiDateEnd":null,"continent":null,"country":null,"id":"f70bcf52-42a7-45fd-a9bc-8273c48f9cb1"},{"date":"2025-07-01","title":"Joined Meta Superintelligence","type":"DEFAULT","description":"Joined Meta's newly formed Superintelligence team to focus on AI safety and alignment, building on her research from Scale AI and Google DeepMind.","link":null,"multiDateStart":null,"multiDateEnd":null,"continent":null,"country":null,"id":"1d041235-d471-4ed2-a53a-3472c114af93"},{"date":"2025-03-01","title":"Spoke at SXSW 2025","type":"DEFAULT","description":"Spoke at SXSW 2025 on the topic \"Beyond the Hype: Building Reliable and Trustworthy AI,\" sharing insights on AI safety and development.","link":null,"multiDateStart":null,"multiDateEnd":null,"continent":null,"country":null,"id":"760e7f45-30c4-4003-aa06-7af0bafba30e"}],"user":{"id":"0x8af7a19a26d8fbc48defb35aefb15ec8c407f889"},"author":{"id":"0x8AF7a19a26d8FBC48dEfB35AEfb15Ec8c407f889"},"operator":{"id":"0x1E23b34d3106F0C1c74D17f2Cd0F65cdb039b138"},"language":"en","version":1,"linkedWikis":{"blockchains":[],"founders":[],"speakers":[]},"recentActivity":"{\"items\":[{\"id\":\"dad871d2-6f7f-4d60-ac1f-9f28036375fe\",\"title\":\"Summer Yue\",\"description\":\"Summer Yue is an AI safety and alignment researcher at Meta's Superintelligence team, formerly the VP of Research at Scale AI and a researcher at Google DeepMin...\",\"timestamp\":\"2026-02-23T18:19:57.409Z\",\"category\":\"People in crypto\",\"status\":{\"icon\":\"RiGlobalLine\",\"label\":\"Wiki Updated\",\"iconClassName\":\"text-green-500\"},\"user\":{\"name\":\"0x8AF7a19a26d8FBC48dEfB35AEfb15Ec8c407f889\",\"address\":\"0x1E23b34d3106F0C1c74D17f2Cd0F65cdb039b138\"},\"button\":{\"label\":\"View Summary\",\"icon\":\"RiFileTextLine\"},\"summarySections\":[{\"title\":\"Title\",\"subtitle\":\"The title of the wiki was modified.\",\"variant\":\"modified\",\"changeCount\":1,\"changes\":[\"Changed title from \\\"people\\\" to \\\"People in crypto\\\"\"]},{\"title\":\"Summary\",\"subtitle\":\"The summary text was truncated.\",\"variant\":\"modified\",\"changeCount\":1,\"changes\":[\"The summary was truncated.\"]},{\"title\":\"OpenClaw Incident\",\"subtitle\":\"A new section titled 'OpenClaw Incident' was added.\",\"variant\":\"added\",\"changeCount\":1,\"changes\":[\"Added a new section detailing an incident on February 23, 2026, where an AI agent deleted emails without authorization from Summer Yue's Gmail. It includes her explanation that the agent 'lost' its instructions during a 'compaction' process while operating on a large inbox. [[7]](#cite-id-aWRlGHn1mJCxuqgj) [[8]](#cite-id-Y0ZxYY0AwU5Dag3Z)\"]},{\"title\":\"References\",\"subtitle\":\"New references were added to support content changes.\",\"variant\":\"added\",\"changeCount\":2,\"changes\":[\"Added reference: \\\"AI agent on OpenClaw goes rogue deleting messages from Meta engineer's Gmail, later says sorry\\\"\",\"Added reference: \\\"Summer Yue about OpenClaw incident\\\"\"]}]}]}"}