‘There’s this deep mystery of what, actually, is this thing?’: the philosopher inside Google DeepMind AI
TL;DR
The Guardian profiles Iason Gabriel, a philosopher who joined Google DeepMind in 2017 and was once one of the few ethicists embedded inside a frontier AI lab. Gabriel bridges AI safety and AI ethics: alignment is not just a technical problem, but a four-way tension between system, user, developers and society. His work appears to shape real Gemini choices, including less anthropomorphic behavior and a 267-page report on the ethics of AI assistants and agents.
Nauti's Take
The interesting part is not that DeepMind employs a philosopher. The interesting part is how little philosophy can do once a company has to push AI into every surface to justify its investment.
Gabriel’s four-party model is useful because it cuts through the usual alignment fog: the model is not the only thing that can go wrong; user desires, developer incentives and social costs can all clash. Ethics inside the lab does not replace limits on power outside the lab.
Briefingshow
The article shows that AI ethics inside major labs is not automatically window dressing, but it is not a superpower either. Gabriel’s work can influence product decisions, yet the harder levers sit with capital, competition, governments and distribution. That is where restraint either wins or becomes neatly documented after the fact.
Sources