- AI companies are increasingly hiring philosophers for AI alignment, safety and governance roles.
- Philosophers help shape AI behaviour, while others study AI consciousness and ethics.
- Experts question how much influence philosophy teams have over commercial AI decisions.
- Companies say human-centred expertise is becoming important as AI adoption expands.
Artificial intelligence companies spent the past few years racing to hire engineers capable of building increasingly powerful models. Now, some of the same companies are recruiting a very different kind of expert.
Google DeepMind recently hired Cambridge philosopher Henry Shevlin to research machine consciousness, human-AI relationships and artificial general intelligence (AGI) readiness. Anthropic employs philosopher Amanda Askell to help shape Claude’s behaviour through the company’s Constitutional AI framework, while OpenAI CEO Sam Altman has said the company consulted “hundreds of moral philosophers” when developing the principles behind ChatGPT.
Meanwhile, a February 2026 report from the Federal Reserve Bank of New York found philosophy graduates now have a lower unemployment rate than computer science graduates. At first glance, it looks as though philosophy has suddenly become one of the AI industry’s most valuable skills.
The reality is more nuanced. AI companies are hiring philosophers, but they are not hiring them to do the same job. Some help define how models should behave. Others study questions around machine consciousness. A third group advises companies on governance and policy, often with varying levels of influence over the products that eventually reach users.
Who Decides How AI Behaves?
The most direct role philosophers play is inside AI alignment teams, where they help answer questions engineers alone cannot.
At Anthropic, Askell leads the personality alignment team responsible for Claude’s Constitution, a document of more than 20,000 words that outlines the principles the model follows when responding to users.
Rather than scripting individual answers, the framework teaches Claude how to balance competing priorities such as safety, legality, honesty and helpfulness across millions of conversations. As AI assistants become more personal, those decisions increasingly involve human behaviour as much as technology.
Maithili Shankar (name changed as requested), an AI engineer from one of the FAANG companies, believes this is where philosophers bring the greatest practical value.
“If we have to include philosophers, it would be mostly in the guardrails team,” Shankar told IBTimes Singapore. “When an AI system becomes too common and too likeable, it’s very easy for a person to start sharing heartbreaks and personal problems with an AI instead of seeking a human being out. It creates a false sense of not being alone while increasing loneliness in the long run.”
As AI increasingly acts as a tutor, assistant and companion, companies are being forced to consider not only what models can answer, but how those answers affect the people asking the questions.
Do Philosophers Really Shape AI?
Not every philosopher hired by an AI company works directly on frontier models. Some researchers focus on machine consciousness and AI welfare. The Financial Times recently reported that Anthropic, Google DeepMind and Meta have expanded work examining whether increasingly capable AI systems could one day deserve some form of moral consideration.
Anthropic has acknowledged studying behaviours resembling panic or anxiety while stressing that it remains deeply uncertain whether such concepts meaningfully apply to AI. Others serve in governance or policy roles, helping companies navigate regulation and responsible deployment.
Whether those roles influence major product decisions remains an open question.
“I’m not really confident whether they are actually taking philosophers into account at all, to be honest,” Shankar said. “Primarily, all these companies, their main aim is revenue generation. There may be people who genuinely want philosophy to improve humanity, but companies still have commercial priorities.”
That concern echoes criticism from Rochester Institute of Technology philosopher Evan Selinger, who has warned that ethics teams risk becoming a form of “ethics washing” if companies promote ethical principles without giving those experts meaningful influence over development decisions.
The distinction matters because only a relatively small number of philosophers work alongside alignment researchers building frontier models. Many others advise from outside the engineering teams.
Why AI Companies Need Philosophers
The role of philosophy inside AI is not limited to ethics. It also extends to understanding how people think, trust and communicate.
Shankar believes philosophical thinking, combined with psychology and neuroscience, can help companies build AI systems that interact with people more naturally.
“They might be harnessing their intelligence,” she said. “Google might hire philosophers and use their understanding of how people think. Philosophy, psychology and neuroscience have influenced people throughout history, and those ideas can influence AI training too.”
She also believes this helps explain why many AI chatbots often appear unusually agreeable. “We don’t like somebody disagreeing with us too much,” she said. “AI models are trained to push back a little when necessary, but otherwise they’ll often agree because that creates a lasting habit. Sometimes you’re essentially getting your own thoughts reflected back to you.”
That tendency may make AI feel more conversational, but it also raises broader questions about whether future assistants should simply satisfy users or challenge them when necessary.
The recent philosophy hires reflect how the AI industry’s priorities are changing. Building larger models remains an engineering challenge. Building systems that billions of people can trust is increasingly becoming a human one.
The title “philosopher” now covers researchers studying machine consciousness, specialists writing behavioural guardrails, and advisers helping companies navigate ethical and regulatory decisions. They may share the same label, but not the same influence.
Whether philosophers ultimately shape the next generation of AI will depend less on their presence inside the world’s biggest labs than on whether they are involved when the decisions that matter most are actually made.
