When people think about artificial intelligence, they usually imagine engineers or programmers building complex systems. But at Anthropic, one of the important people shaping how AI behaves is actually a philosopher. Amanda Askell is helping teach AI systems how to respond to people in a responsible and ethical way.
A Philosopher working with AI
Amanda Askell is a philosopher and AI researcher who works at Anthropic. She leads a team that focuses on how the company’s AI chatbot, Claude, interacts with users.
Her work is not about writing code alone. Instead, she helps decide how AI should behave when answering questions, especially difficult or sensitive ones. The goal is to make sure the AI remains helpful, honest and safe while talking to people.
Creating rules for AI behaviour
One of Askell’s key contributions is helping create the guiding principles for Claude. These principles act like a “constitution” that helps the AI understand how it should respond to different situations.
Instead of following only strict rules, the system is trained to follow broader values such as honesty, fairness and responsibility. This approach helps the AI handle new situations better and reduces the chances of harmful or misleading responses.
From Philosophy to Technology
Before joining Anthropic, Askell worked at OpenAI where she studied AI safety and ethics. She later became one of the early members of Anthropic when the company was founded in 2021.
Her career shows that building advanced AI is not only about technology. Experts from fields like philosophy and ethics are also becoming important in shaping how AI systems work.
Why ethics in AI matters?
As AI tools become more common in everyday life, companies are paying more attention to how these systems behave. Experts believe that AI should not only be powerful but also responsible.
Amanda Askell’s work highlights this shift. By combining philosophy with technology, she is helping ensure that modern AI systems are designed to act in ways that respect human values.
Meanwhile, Tom Blomfield, a partner at Y Combinator, recently made a comment on social media that has reignited debate about how AI might change professional services. According to Blomfield’s post, “a 24-year-old who learned Claude Code last Tuesday is about to outperform the entire Accenture workforce.” His remark highlights the growing notion that strong AI systems, such as tools like Claude from Anthropic, might enable one skilled worker to finish tasks that previously required large teams.