PeopleofSuperintelligence

Lab leader

Jason Stanley

Jason Stanley leads Applied AI Research at ServiceNow, where his work spans AI policy, language models and agent security. He was elected a Canadian Rhodes Scholar in 2000; his education includes a PhD in sociology from New York University and social-science degrees from Oxford University and Williams College. At Local Logic in 2021, he discussed how land-use decisions shape transportation patterns and location sustainability; in 2024, he coauthored LitLLMs on language-model-assisted literature reviews. His 2025 papers brought social frameworks into LLM alignment, tested security risks for AI agents and studied fine-tuning attacks. By 2025, he headed AI Research Deployment at ServiceNow; in October 2026, he discussed oversight for AI systems that continue to adapt after deployment.

Works
4

Publication numbers from OpenAlex, as of 22 October 2026. Citations and h-index are hidden because this OpenAlex record may be incomplete or mix in other authors.

Timeline

  1. 2000

    A Rhodes Scholarship

    He is elected a Canadian Rhodes Scholar in 2000 in the Newfoundland and Hertford constituency. He holds a PhD in sociology from New York University and social-science degrees from Oxford University and Williams College.

  2. 2021

    Head of Insights at Local Logic

    In 2021, he is Head of Insights at Local Logic and explains how land-use decisions shape transportation patterns and location sustainability. He connects property choices with emissions from travel to and from each asset.

  3. 2024

    LLMs for literature reviews

    In 2024, he coauthors LitLLMs, which studies how LLMs retrieve related work from an abstract and draft literature reviews from the results. The paper builds rolling evaluation sets from arXiv papers to reduce test contamination as new LLMs appear.

  4. By 2025

    AI research deployment

    By 2025, he heads AI Research Deployment at ServiceNow, where his team de-risks AI research and seeks product value. His earlier work includes applied research on AI trust and governance and research and product leadership at technology companies. His policy work includes an invited expert role on AI issues for the OECD, service on the Partnership on AI’s Expert Group on Human-AI Collaboration, and labor-market policy work for the Government of Canada.

  5. February 2025

    Fine-tuning safety attacks

    A 2025 study he coauthors finds that fine-tuning attacks can train models to refuse harmful requests before answering, bypassing token-level safety mechanisms.

  6. February 2025

    Societal alignment frameworks

    In 2025, he coauthors Societal Alignment Frameworks Can Improve LLM Alignment, which argues for using social, economic and contractual frameworks to inform LLM alignment.

  7. April 2025

    DoomArena security testing

    DoomArena is a 2025 security-evaluation framework for AI agents that he coauthors to test evolving security threats.

  8. As of October 2026

    Applied AI Research at ServiceNow

    As of October 2026, he is Head of Applied AI Research at ServiceNow. In an October 2026 episode of The AI Drop, he discusses oversight for AI systems that continue to adapt after deployment.

Career

  1. NowServiceNowHead of AI Research Deployment
  2. From 2021Local LogicHead of Insights
  3. Organisation for Economic Co-operation and DevelopmentInvited expert on AI issues
  4. Partnership on AIExpert Group on Human-AI Collaboration member

Education

  1. New York UniversityPhD
  2. University of Oxford
  3. Williams College

From the biography.

  1. Direct Form 990 pay in the latest returns of 39 nonprofits with AI people on this site is highest at $2.1M at the Allen Institute for AI (Ai2) (2024)Includes Partnership on AI. IRS Form 990 compensation across 39 nonprofits with AI people on this site from 2012 to 2025: in the latest returns, Mozilla reports $6.1M including related-company pay (2024) and the Allen Institute for AI (Ai2) leads direct nonprofit pay at $2.1M (2024).
  2. The company share of AI researchers’ top papers was 31% in 2020–2024 and 8% in 1970–1974Includes Partnership on AI. We estimate where the people shaping AI worked when their most-cited papers appeared: the company share was 31% in 2020–2024 and 8% in 1970–1974, and Google led all organizations in 2020–2024.
  3. Google has the highest combined h-index among AI organizationsIncludes Partnership on AI. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  4. The median h-index is 48 at universities and 14 at frontier AI labsIncludes University of Oxford. The median h-index is 48 at universities and 14 at frontier AI labs. We compare publishing careers, recent citations, dual appointments and job moves across 2,420 university researchers and 308 frontier-lab researchers.

Sources and credits

Photo: We have no published photo of Jason Stanley yet. Send us a free-licensed photo.

  1. Jason Stanley | Rhodes Trust rhodeshouse.ox.ac.uk. Checked 21 October 2026.
  2. Location Sustainability and the Future of Cities | Local Logic locallogic.co. Checked 21 October 2026.
  3. [2412.15249] LitLLMs, LLMs for Literature Review: Are we there yet? arxiv.org. Checked 21 October 2026.
  4. Jason Stanley - frontiers-in-service-2025.hec.ca frontiers-in-service-2025.hec.ca. Checked 21 October 2026.
  5. [2502.19537] No, of Course I Can! Deeper Fine-Tuning Attacks That Bypass Token-Level Safety Mechanisms arxiv.org. Checked 21 October 2026.
  6. [2503.00069] Societal Alignment Frameworks Can Improve LLM Alignment arxiv.org. Checked 21 October 2026.
  7. [2504.14064] DoomArena: A framework for Testing AI Agents Against Evolving Security Threats arxiv.org. Checked 21 October 2026.
  8. The AI Drop #25: Governing Adaptive AI Agents | Freshfields freshfields.com. Checked 21 October 2026.