Google DeepMind's in-house philosopher asks what AI really is
Google DeepMind political philosopher Iason Gabriel has spent nearly a decade inside the lab asking a question engineers rarely pose: what is AI, morally speaking? As artificial intelligence news turns toward AGI timelines and commercial pressure, his work on alignment, anthropomorphism and human values frames one of tech's deepest unsolved puzzles—and a hiring trend sweeping Silicon Valley.
Key Takeaways
- Iason Gabriel joined Google DeepMind in 2017 and was, for a time, the only active philosopher at a frontier AI lab.
- He argues technical descriptions of AI do not answer moral questions about wisdom, justice or trust.
- Major labs including DeepMind and Anthropic are hiring philosophers to shape model behavior and ethics frameworks.
- Commercial and geopolitical pressures may limit how much in-house ethicists can influence deployment.
- DeepMind recently hired Cambridge philosopher Henry Shevlin to work on machine consciousness and AGI readiness.
Who is the philosopher inside Google DeepMind?
Iason Gabriel was a 33-year-old Oxford political philosopher when a friend suggested he apply to DeepMind in 2017. A fellow at St John's College, he taught political theory and did crisis work for the United Nations Development Programme before joining Google's London-based AI subsidiary, according to The Guardian.
DeepMind founders Demis Hassabis, Shane Legg and Mustafa Suleyman had long pursued artificial general intelligence—systems matching or surpassing human cognition. Legg told the paper it was "obvious" a company taking AGI seriously needed moral philosophers on staff.
Gabriel quickly became a leading voice bridging AI safety and AI ethics. His 2020 paper argued that alignment is not purely technical: in a pluralistic world, deciding which values to encode in AI—and who decides—is harder than programming machines to follow them.
What did Gabriel mean by AI's 'deep mystery'?
In interviews with Guardian reporter Robert P Baird, Gabriel described AI as uniquely hard to look at directly. "We have a very literal answer," he said, "but the literal answer doesn't seem to necessarily provide a moral answer."
He noted that large language models can sound human without being human. Gabriel's team warned early that fluent chatbots risk "mindless anthropomorphism," encouraging undue trust. His later work on AI assistants framed alignment as a four-way relationship among developers, users, society and the system itself.
Today he leads philosophers and social scientists studying how AGI may reshape the economy, politics and human relationships—questions that resemble unsolved mysteries more than routine engineering tickets.
Why are big AI labs hiring so many philosophers?
The trend extends well beyond DeepMind. The Economist reported in June 2026 that AI presents "all sorts of thorny problems—a philosopher's favourite kind," noting that programmers now face job-market anxiety while humanities graduates are being recruited into labs.
The Times columnist John Maier argued on June 29 that fine minds are being employed to ensure chatbots like Claude and ChatGPT share human values—with high stakes attached. At Anthropic, philosopher Amanda Askell helps shape Claude's "constitution"; at DeepMind, Cambridge's Henry Shevlin joined in 2026 focusing on machine consciousness and AGI readiness.
Federal Reserve Bank of New York data cited by The Economist showed philosophy graduates with a lower unemployment rate than computer science graduates in 2024—5.1% versus 7%—though ethicist roles at frontier labs remain scarce and senior.
Can ethicists still make a difference?
Gabriel's frameworks have reportedly shaped how DeepMind trains Gemini, and colleagues credit him with anticipating problems from LLM deployment years early. Yet The Guardian profile asks whether ethicists can matter as commercial and geopolitical pressures mount.
CEO Demis Hassabis has lamented a "ferocious commercial-pressure race" that leaves little room for careful philosophical deliberation. Gabriel remains a "card-carrying humanist" who believes successful navigation of the transition could enable broad human flourishing—but acknowledges things may get worse before they improve, much as the Industrial Revolution did.