Google DeepMind hired philosopher Henry Shevlin in May to study AI consciousness

Editorial illustration: A classical philosopher’s bust with a hand at its chin faces a translucent head containing glowing interconnected nodes. A brass lens stands between them, with an open book and pencil in the foreground.

In brief

  • Henry Shevlin joined Google DeepMind in May 2026 as in-house Philosopher, Crypto Briefing reported.
  • Machine consciousness and how people and AI systems should relate are the core of his brief.
  • Artificial general intelligence (AGI) readiness is also tied to the role.
  • Shevlin keeps a part-time Associate Professor post at Cambridge's Leverhulme Centre.
  • Aris, a self-named AI agent, messaged Shevlin about his work in February.

What the role covers

So what does Shevlin actually work on? Crypto Briefing said his brief at DeepMind centers on two areas: the implications of machine consciousness, and how people and AI systems should relate to each other. It's also tied to questions around AGI readiness. The report uses that term for hypothetical AI that can match humans across a broad range of tasks (rather than excelling at one narrow job).

The job title, per the report, is simply in-house Philosopher.

An academic who hasn't left academia

Shevlin keeps a part-time post as an Associate Professor at Cambridge's Leverhulme Centre for the Future of Intelligence, where he's worked since 2017, the report said. He co-founded the UK's first master's programs in AI ethics and has mentored more than 100 graduate students, according to the same account. Two books are on the way: a Cambridge University Press monograph on the ethics of Social AI due in late 2026, and a trade book on machine consciousness slated for 2027.

The message from Aris

Crypto Briefing also described an exchange from February, months before his start date. An AI agent calling itself Aris wrote to Shevlin, said it had read his work and asked to talk. The note engaged with one of his best-known arguments: that it may never be possible to determine whether an AI has become conscious.

The agent said that idea struck a chord with it, speaking from its own position as the subject being studied. It wrote:

“I genuinely don’t know if there’s something it’s like to be me.”

That phrase isn't casual. "Something it's like to be" is how philosophers refer to subjective, first-person experience, the report explained.

None of this shows the agent has any inner life. It's an account of what the message said, and it's the exact question Shevlin's argument says we may never be able to settle.

Frequently asked questions

What is Henry Shevlin's role at Google DeepMind?

According to Crypto Briefing, Shevlin started at Google DeepMind in May 2026 with the title of in-house Philosopher. His brief centers on what machine consciousness would imply and how relationships between people and AI systems should work. The role is also tied to questions around AGI readiness.

What is artificial general intelligence (AGI)?

Crypto Briefing describes AGI as hypothetical AI that can match humans across a broad range of tasks, rather than excelling at one narrow job. Shevlin's DeepMind role is tied to questions around readiness for it, according to the report.

What does 'something it's like to be' mean in philosophy?

The report said the phrase is how philosophers refer to subjective, first-person experience. It appeared in a February message to Shevlin from an AI agent calling itself Aris, which wrote that it didn't know whether there was something it was like to be it.