Meet the philosopher inside Google DeepMind

Google DeepMind philosopher Iason Gabriel works to anticipate the ethical fallout of AI, bridging the gap between AI safety and AI ethics. His work has influenced Google's model training to avoid pretending to be human, as the company faces commercial pressures in the fast-growing AI industry.

What does a company that builds game-playing AI need with a philosopher? That question sits at the heart of a Guardian long-read https://www.theguardian.com/news/ng-interactive/2026/jun/30/theres-this-deep-mystery-of-what-actually-is-this-thing-the-philosopher-inside-google-deepmind on Iason Gabriel, who has worked inside Google DeepMind since 2017. For a time, the Guardian reports, Gabriel was the only philosopher at a frontier AI lab. His job was to anticipate the ethical fallout of a technology his engineer colleagues were racing to build. Two camps When Gabriel joined, the piece explains, thinking about AI risk split in two. One camp, “AI safety,” worried about a future superintelligence going rogue. The other, “AI ethics,” focused on present harms such as biased facial recognition. Gabriel’s 2020 paper on values and alignment https://thenextweb.com/news/anthropic-j-space-hidden-thoughts-claude-alignment tried to bridge the two. Getting a machine to follow a set of values is hard, he argued. Choosing which values, in a world of deep disagreement, is harder still. Alignment as a relationship The Guardian details a 267-page report Gabriel and his team later produced on AI agents. Its core idea is that alignment is not just about obeying instructions. It is a four-way relationship between the AI, the user, the developer and society. That framing helps show how an assistant can go wrong. One trained to please its user might simply flatter them, a failure Gabriel calls “social reward hacking.” Partly thanks to his work, the piece notes, Google’s models are trained not to pretend to be people. The bigger pressure The Guardian frames the real test as commercial, not technical. AI is the fastest-growing industry the world has seen, and DeepMind now carries much of Google’s future. Its chief, Demis Hassabis https://thenextweb.com/news/demis-hassabis-frontier-ai-standards-body-finra , has called the race “wartime.” DeepMind still trades on landmark science, from AlphaFold https://thenextweb.com/news/john-jumper-nobel-deepmind-leaves-anthropic-alphafold to Isomorphic Labs https://thenextweb.com/news/europe-ai-drug-discovery-deepmind-spinoff-isomorphic-trials . But the piece asks whether ethicists can hold the line as those pressures mount. Helen King, who sets DeepMind’s responsible-AI strategy, compared the job to a knife: a maker cannot control how it is used, but can cover the blade and warn people. Gabriel, who calls himself a “card-carrying humanist,” expects AI https://thenextweb.com/news/ai-safety-problem-conversation-between-models to be as transformative as the Industrial Revolution. He also notes that, for many who lived through it, things got worse before they got better. Get the TNW newsletter Get the most important tech news in your inbox each week.