PeopleofSuperintelligence

Researcher

Amanda Askell

Amanda Askell is a philosopher and AI researcher who works on model behavior at Anthropic. Her academic path includes Dundee, Oxford, and a New York University doctorate on infinite ethics. She previously worked on AI policy at OpenAI. Her research contributions include work on a general language assistant, InstructGPT, and Constitutional AI with many colleagues. These projects examine how instructions, feedback, and written principles can shape a model's responses. Anthropic sets out aims for how Claude should behave. Askell is the primary author of Anthropic’s 2026 constitution, which credits other contributors. The documents describe training intentions and desired behavior. They leave actual reliability and consistency to be tested in use.

Now at
Anthropic
Works
33

Publication numbers from OpenAlex, as of 14 October 2026. Citations and h-index are hidden because this OpenAlex record may be incomplete or mix in other authors.

Photo of Amanda Askell

Timeline

  1. University training

    Beginning with philosophy

    Askell studied philosophy at Dundee and Oxford, then earned a doctorate at NYU.

  2. 2021

    An assistant as a research setting

    At Anthropic, Askell co-authors A General Language Assistant as a Laboratory for Alignment with a large team.

  3. 2022

    Learning from human instructions

    She is also a co-author of the InstructGPT paper, which studies demonstrations, response rankings and reinforcement learning.

  4. December 2022

    Training with stated principles

    Askell co-authors Constitutional AI with Yuntao Bai and many colleagues.

  5. 2026–today

    Writing a fuller constitution

    Askell is the primary author of Anthropic’s new constitution, with important contributions from Joe Carlsmith, Chris Olah and others.

Sources and credits

Photo: Amanda Askell; image supplied by askell.io. License: Not stated by source. Source: askell.io. Cropped to a square and resized. Ask us to remove or replace this photo.

  1. Amanda Askell: personal biography askell.io. Checked 10 October 2026.
  2. Askell: Pareto Principles in Infinite Ethics askell.io. Checked 10 October 2026.
  3. Askell and colleagues: general assistant laboratory, December 2021 arxiv.org. Checked 10 October 2026.
  4. Ouyang and colleagues: InstructGPT, March 2022 arxiv.org. Checked 10 October 2026.
  5. Bai and colleagues: Constitutional AI, December 2022 arxiv.org. Checked 10 October 2026.
  6. Anthropic: Claude’s Character, June 2024 anthropic.com. Checked 10 October 2026.
  7. Anthropic: Claude’s Constitution, January 2026 anthropic.com. Checked 10 October 2026.