Ex-Google DeepMind researcher Bilal Chughtai warns AI could kill us all

The warning from Bilal Chughtai is the latest in a series of public concerns from people who have worked inside major AI labs. Chughtai says he worked on AGI safety and alignment at Google DeepMind before leaving in July, and his Monday post on X follows comments from Anthropic researchers Jacob Coxon and Evan Hubinger. That pattern—insiders at several leading labs raising existential risk—is what gives the remarks their weight beyond any single post.

The body frames the debate as a split over whether the race toward increasingly capable AI should slow. Anthropic CEO Dario Amodei has called for slower advanced-AI development, with support from SpaceX CEO Elon Musk and OpenAI CEO Sam Altman. President Donald Trump has pushed the opposite way, dismissing calls for greater regulation as a hoax. For Alphabet, the body says the biggest market risk is not the researchers' worst-case scenario itself but whether safety concerns translate into regulation, slower model releases, or higher development costs.

What happens next hinges on whether the warnings begin changing policy or corporate behavior rather than simply intensifying public debate. If they do, tighter restrictions could affect companies supplying compute and infrastructure when model-development timelines stretch. If regulatory action stays limited, the current AI spending race would likely remain largely intact, leaving Alphabet's near-term earnings picture less directly exposed to the safety debate itself.