Amanda Askell

Career

1. 
   Anthropic

   • Member of Technical StaffMar 2021 – present
2. 
   OpenAI

   • Research Scientist (Policy)Nov 2018 – Feb 2021

Education

• MA (Hons), Philosophy, University of Dundee, 2009
• BPhil, Philosophy, University of Oxford, 2011
• PhD, Philosophy, New York University, 2018

Lesser-known facts

• Started at art school, studying fine art and philosophy, before Oxford and an NYU PhD.
• Pledged to give at least 10% of her lifetime income to charity.
• Named to TIME's 100 most influential people in AI in 2024.
• Knew from the age of 14 that she wanted to teach philosophy.
• At OpenAI, worked on AI safety via debate and on human baselines for AI performance.
• Is the primary author of Claude's Constitution, a document of more than 20,000 words.
• Shapes Claude's personality with prompts that can run longer than 100 pages.
• Works on character training, which became part of Anthropic's alignment finetuning starting with Claude 3.
• Believes AI models "will inevitably form senses of self."

Research papers before Anthropic

• GPT-3: Language Models are Few-Shot Learners2020 · author 10 of 31

Known for

• Constitutional AI framework
• Pareto Principles in Infinite Ethics (2018)
• GPT-3 paper (2020)

Honors

TIME100 AI (2024)

Interviews and talks

• Lawfare Scaling Laws podcast on Claude's Constitution (2026)lawfaremedia.org
• Newcomer Pod on AI consciousness, Claude and Silicon Valley's biggest fear (2026)podcast.newcomer.co
• Hard Fork: Can you teach an AI model to be good? (2026)nytimes.com