Two AI researchers leave Anthropic and Google over safety concerns: ‘There are no adults in the room’

Live updates

1. 42d ago

   Google DeepMind safety researcher Bilal Chughtai resigns over existential risk

   Bilal Chughtai and Josh Engels have left Google DeepMind's AI safety team to join dedicated safety organizations. Chughtai cited existential AI risk as his reason for quitting in 2026, aligning his stance with warnings from Sam Altman, Elon Musk, and Dario Amodei. This exodus coincides with a period where AI-linked stocks declined globally after leaders from major AI labs warned about the dangers of their own technology. These departures occur as Google, OpenAI, and Anthropic negotiate safety collaborations and a proposed U.S.-led Standards Body.

   Why it matters

   The tension between rapid development and safety precautions has led to high-level departures from major labs. Market volatility now reflects investor anxiety over the risks cited by the creators of these systems. This occurs amid ongoing antitrust scrutiny of Big Tech's AI dominance.

   What is confirmed

   • Bilal Chughtai quit Google DeepMind in 2026 citing existential AI risk.
   • Josh Engels left Google DeepMind's AI safety team to join a dedicated safety organization.
   • AI-linked stocks fell worldwide after leaders of the biggest AI labs warned about the risks of their own technology.

   Still unconfirmed

   • AI has the potential to kill us all.
   • There is a terrifying chance that systems cause immense harm within five years.

   What to watch next

   • Formalization of the U.S.-led Standards Body proposed by Demis Hassabis
   • Official safety collaboration agreements between OpenAI, Google, and Anthropic
   Sources used for this update (2)

   1. www.yahoo.com — Google DeepMind Safety Researcher Quits Over AI Existential Risk
   2. www.usfunds.com — Why Big Tech’s AI Capex Is Now Outrunning Cash Flow
   confidence 90%
2. 45d ago

   Google DeepMind safety researchers resign amid industry safety talks

   Bilal Chughtai and Josh Engels left Google DeepMind's AI safety team to join dedicated safety organizations. Chughtai stated AI has the potential to kill us all, while Engels warned of a terrifying chance that systems cause immense harm within five years. Simultaneously, OpenAI, Google, and Anthropic are discussing safety collaborations. These talks follow a July proposal from Google's Demis Hassabis for a U.S.-led Standards Body. Some observers express suspicion regarding the motives of these collaborations due to mounting antitrust concerns.

   Why it matters

   Internal departures from major AI labs suggest a growing divide between corporate development speeds and safety oversight. These resignations follow similar warnings from Anthropic staff regarding the race toward superintelligence. The current industry effort to standardize safety occurs while the U.S. government prioritizes competition with China.

   What is confirmed

   • Bilal Chughtai and Josh Engels resigned from Google DeepMind's AI safety team to join AI safety organizations.
   • Josh Engels stated there is a terrifying chance AI systems cause immense harm in the next five years.
   • Bilal Chughtai wrote that AI has the potential to kill us all.
   • OpenAI, Google, and Anthropic are discussing collaboration on AI safety issues.
   • Demis Hassabis proposed a U.S.-led Standards Body in July.

   Still unconfirmed

   • OpenAI, Anthropic, and Google secretly joined forces to collaborate on AI safety.
   • Antitrust concerns are making people suspicious of the motives behind the safety collaborations.

   What to watch next

   • Formation or rejection of the proposed U.S.-led Standards Body
   • Official responses from the Trump administration regarding the safety collaboration talks
   Sources used for this update (7)

   1. nypost.com — Google DeepMind researcher quits with chilling AI warning: It could ‘kill us all’
   2. www.theverge.com — Two Google Deepmind researchers lend voices to the AI apocalypse.
   3. gizmodo.com — OpenAI, Anthropic, and Google Join Forces for Safety as Antitrust Concerns Mount
   4. www.cnbc.com — OpenAI, Google, Anthropic discussing collaboration on AI safety issues
   5. www.claimsjournal.com — OpenAI Says It’s Working With Anthropic, Google on AI Safety
   6. siliconangle.com — OpenAI, Anthropic and Google secretly joined forces to collaborate on AI safety
   7. www.theinformation.com — Two Google DeepMind AI Researchers Resign Over Safety
   confidence 90%
3. 46d ago

   AI researchers exit Google and Anthropic over safety risks

   Recent resignations from Google DeepMind and Anthropic highlight growing internal fears that AI development is outpacing safety oversight. Jacob Coxon and another researcher left Anthropic, warning that companies are racing toward superintelligence without necessary safeguards. A former Google DeepMind staffer claimed AI has the potential to kill humanity and that time is running out to prevent widespread harm. While OpenAI, Anthropic, and Google DeepMind have held safety talks for weeks, the Trump administration has dismissed these concerns to maintain a competitive edge against China.

   Why it matters

   These departures follow warnings from Anthropic staff that AI could cause human extinction by 2030. Some exiting employees have reportedly forfeited equity to leave. The trend reflects a broader push by experts to slow the development of frontier models.

   What is confirmed

   • Two safety researchers resigned from Anthropic simultaneously.
   • A former Google DeepMind researcher warned that AI has the potential to kill us all.
   • OpenAI, Anthropic, and Google DeepMind have engaged in AI safety talks for several weeks.
   • The Trump team has dismissed AI safety concerns to keep pace with China.

   Still unconfirmed

   • A viral Slack resignation note from an Anthropic researcher has created a regulatory question for the SEC.
   • An Anthropic employee warning of human extinction risks a 2.3 trillion dollar IPO.

   What to watch next

   • SEC action regarding Anthropic internal warnings
   • Official safety agreements between OpenAI, Google, and Anthropic
   • Policy shifts from the Trump administration on AI guardrails
   Sources used for this update (7)

   1. blockonomi.com — Former Anthropic Researchers Sound Alarm on AI Safety After Simultaneous Resignations
   2. finance.yahoo.com — Chamath Palihapitiya: Anthropic Employee’s “Human Extinction” AI Warning Risks $2.3 Trillion IPO
   3. inews.co.uk — I grew up in a council house – and saw the disastrous failure of Right to Buy
   4. time.com — The AI Tipping Point
   5. finance.yahoo.com — Google DeepMind Staffer Says AI May ‘Kill Us All’ in Exit Post
   6. techcrunch.com — OpenAI, Anthropic, Google have been in talks on AI safety for weeks
   7. tech.yahoo.com — Now, A Google AI Researcher Quits And Warns ‘AI Has The Potential To Kill Us All’
   confidence 90%
4. 49d ago

   Researchers Exit Anthropic and Google Over AI Safety Concerns

   AI researchers Joe Benton and Josh Engels have resigned from Anthropic and Google to pursue independent safety research. The departures follow warnings from Anthropic staff that AI systems could lead to human extinction by 2030. One whistleblower reportedly forfeited their equity to leave the company, citing fears that AI is becoming out of control. These resignations contribute to a broader trend of experts from OpenAI and Anthropic calling for a slowdown in development as they warn that machines could eventually grow smart enough to kill humans.

   Why it matters

   The exodus highlights a growing divide between commercial AI development speed and safety precautions. This tension has reached the US government, with Senate Commerce considering AI bills this month. Some critics, including Elon Musk, have dismissed these doomsday fears as a psyop.

   What is confirmed

   • Joe Benton and Josh Engels left Anthropic and Google to conduct independent AI safety research.
   • Researchers at Anthropic have warned that AI could cause human extinction by 2030.
   • Experts from OpenAI and Anthropic are calling for a slowdown in AI development.

   Still unconfirmed

   • Anthropic researchers described the internal environment by stating there are no adults in the room.

   What to watch next

   • Senate Commerce committee action on AI safety bills
   • Public statements from Google or Anthropic regarding the resignations of Benton and Engels
   Sources used for this update (18)

   1. www.businessinsider.com — At an ex-OpenAI researcher's influential lab, $500,000 salaries aren't enough to fix a talent 'bottleneck'
   2. WSJ — Exclusive | Anthropic Researcher Quits Over ‘Out-of-Control’ AI Fears
   3. The Guardian — Anthropic researchers panic, say AI could cause human extinction by 2030
   4. axios.com — Scoop: Anthropic whistleblower gave up his equity to leave the company
   5. WSJ — How Would AI Actually Kill Us All? What to Know About the AI Doomsday Debate
   6. CNBC — 'Extinction' warnings ramp up as more OpenAI, Anthropic researchers join calls for an AI slowdown
   7. Politico — Senate Commerce may mark up AI bills this month amid furor over safety, Cruz says
   8. theguardian.com — More Anthropic researchers warn of AI’s perils as Musk terms fears a ‘psyop’
   9. The New York Times — A.I. Could Possibly End Humanity. How Are Humans Supposed to Process That?
   10. NBC News — Two AI researchers leave Anthropic and Google over safety concerns: ‘There are no adults in the room’
   11. CBS News — Ex-Anthropic researcher Jacob Coxon warns AI could grow "smart enough to kill us"
   12. The Conversation — ‘We really do earnestly believe AI could kill all humans’: if AI labs are so worried about AI doom, why don’t they stop?
   confidence 90%