Ex-Anthropic staffer claims AI risks are binary in new post
According to Alignment Alignment, a former Anthropic employee with $15 million in stock options argued that AI outcomes are binary—either extinction or none—dismissing inequality harms as unlikely, as highlighted by Timnit Gebru.
SourceAnalysis
On October 15, 2026, researcher Timnit Gebru posted a comment highlighting statements from a former Anthropic employee who holds $15 million in company stock options, claiming that the only serious AI possibilities worth considering are whether the technology will kill all humans or not.
The post, referencing a blog at alignmentalignment.ai, mocks the dismissal of other outcomes such as AI making things worse or increasing inequality as unlikely or impossible, drawing attention to debates over AI risk framing within the industry.
timnitGebru (@dair-community.social/bsky.social)
@timnitGebruAuthor: The View from Somewhere Mastodon @timnitGebru@dair-community.