This image is not related to the article. [123RF]
This image is not related to the article. [123RF]

Bilal Chughtai, a researcher who recently left Google DeepMind, warned that he "genuinely believes AI has the potential to kill us."

Chughtai made the remarks Monday in a post on X, formerly Twitter, saying, "At Google, I witnessed AI development firsthand. I am also extremely concerned about the fundamental direction of this technology," Yonhap reported.

Chughtai said he had worked on artificial general intelligence safety and alignment research — work focused on training AI models to refuse actions that could pose a threat to humans.

Leaders of 4 AI firms unusually unite behind calls to slow development

Earlier, the heads of four companies that have fiercely competed in AI development — Anthropic, OpenAI, SpaceX's xAI and Google DeepMind — unusually spoke with one voice, agreeing that AI development should be slowed.

Dario Amodei, CEO of Anthropic, compared the pace of AI progress to a curve following the sequence "1, 2, 4, 8, 16, 32," saying the curve is now "starting to steepen," in an interview aired Monday on CBS News, Yonhap reported.

"There's no need to panic right now, and there's no need to shut everything down," he said, adding, "Even so, this is a warning signal that we need to slow down."

OpenAI CEO Sam Altman also weighed in on X, writing, "I agree with Dario that we should be moderating the pace at the frontier."

Elon Musk, CEO of SpaceX, shared Amodei's post on X and wrote, "Dario is right."

Demis Hassabis, CEO of Google DeepMind, said on X that "Dario's essay is pointing in the right direction," adding, "The details need to be refined, but it's the right direction for this critical moment."

Amodei's remarks came just days after Anthropic researcher Jacob Coxon resigned Wednesday, warning that AI could threaten humanity and saying he did not want to work in an industry he considered "out of control."


yul@heraldcorp.com