The possibility that we may have moral obligations to AI systems is receiving increased attention by academics and frontier AI labs. For example, Anthropic hired an AI welfare researcher late in 2024.

This question resolves YES if Google DeepMind is known by publicly available information to have hired someone whose primary role is researching AI welfare. Ideally it is publicly reported that a specific person has been hired in a role with the term "AI welfare" in it, like AI welfare researcher or AI welfare expert or AI welfare associate, but any role that is substantively and primarily about AI welfare research (with or without a specific named person, as long as we know someone is hired) would be sufficient to resolve this question to YES.

Otherwise, this question resolves NO on 1 Jan 2026.

See also: