← Back to Stories

AI researchers warn of existential risks from superintelligence, call for urgent regulation

By Updated 3 hours ago3 articles from 3 independent sources

Consensus Summary

Multiple AI researchers, including Jacob Coxon, a 28-year-old former Anthropic and OpenAI employee, have resigned or spoken out this week warning that artificial superintelligence could pose an existential threat to humanity. Coxon and others at Anthropic, including Evan Hubinger, stated there is a greater than 10% chance AI could 'kill all humans' by the end of the decade, prompting calls for urgent global regulation. The warnings came as MPs in Westminster convened a session on Monday to discuss AI risks, with figures like Beatrice Fihn, a 2017 Nobel peace prize winner, emphasizing the need for international controls. Meanwhile, over 1,300 AI employees signed a July statement urging governments to develop tools to 'pace' AI development, and politicians like Bernie Sanders have pushed for legislative action. Concerns about AI alignment, safety, and unchecked development have intensified, with some researchers arguing that current efforts lack responsible oversight.

✓ Verified by 2+ sources

Key details reported by multiple sources:

  • Jacob Coxon, a 28-year-old researcher at Anthropic (and previously at OpenAI), resigned and claimed 'neither company was acting responsibly' regarding AI safety
  • Anthropic employees (including Evan Hubinger) stated there is a greater than 10% chance AI could 'kill all humans' in the next decade
  • A Westminster session on Monday was convened by Control AI to discuss AI regulation, with MPs and peers attending
  • Beatrice Fihn, who won the 2017 Nobel peace prize, warned about AI risks during the Westminster session
  • Anthropic’s latest model, Mythos 5.1, has had limited access, with only a few US organizations reviewing it
  • More than 1,300 employees from frontier AI companies signed a statement in July calling for international tools to 'pace' AI development
  • AI regulation bills were introduced in parliament this week, including one by Labour MP Alex Sobel to ban artificial superintelligence
  • Bernie Sanders called for AI regulation this week, citing 81% of Americans believing politicians should act

Points of Difference

Details reported by only one source:

The Guardian
  • Control AI is funded by Jaan Tallinn, who estimates 10-15% of AI employees believe AI will be a worthy successor to humanity
  • Tallinn said an AI researcher told him 'Humans are a disposable species' earlier this summer
  • Former defence secretary Des Browne compared AI risks to nuclear threats
  • Labour MP Darren Jones wrote to the prime minister and UN/OECD calling for a 'multinational treaty' on AI safety
  • Dr Andrew Rogoyski and David Barber expressed cautious optimism about AI's limitations and controllability
  • Sandra Wachter and Gary Marcus discussed AI risks like misinformation and job displacement, calling 'Terminator scenarios' a distraction
Sydney Morning Herald
  • Jacob Coxon’s resignation post was viewed more than 100 million times
  • Anthropic’s Samuel Marks stated AI extinction risks could happen in the next few years, with senior employees more concerned
  • US President Donald Trump framed AI regulation as a burden that could give China an advantage
  • Florida Republican Ron DeSantis and Illinois Democrat JB Pritzker voiced concerns, with DeSantis noting potential investor-driven hysteria
  • Anna Paulina Luna called for a special session of Congress on AI, emphasizing bipartisan urgency
ABC News
  • UN rights chief Volker Turk warned AI could pose an existential threat and called for 'cast-iron guarantees' on AI safety
  • Anthropic’s CEO Dario Amodei was mentioned in response to the 'Pacing the Frontier' initiative
  • OpenAI’s chief scientist Jakub Pachocki stated last month that international coordination on AI development is needed

Where the reporting differs

Details that conflict, or appear in only some outlets:

  • The Guardian and SMH mention AI risks being discussed 'this week,' but ABC does not specify a timeline for the UN’s Volker Turk’s warning
  • The Guardian and SMH refer to AI risks as 'by the end of the decade,' while ABC does not explicitly state a timeline for extinction risks beyond Coxon’s 'end of the decade' claim
  • The Guardian mentions a 'Tuesday night' admission by an Anthropic employee, but SMH and ABC do not specify this exact timing

Source Articles

GUARDIAN

Will it kill off humanity? Computer scientists and politicians wrestle with risks of AI

Alarming warnings from industry insiders increase pressure for curbs on artificial superintelligence Does artificial superintelligence really pose a risk greater than nuclear weapons? Is there a significant chance of “a Chornobyl-sized catastrophe”. Might there even be a greater than 10% chance that AI could “kill all humans” in the next decade? These warnings were issued over the past 48 hours on both sides of the Atlantic about the potential impact of a technology that most people still think

SMH

‘Could kill us all’: Anthropic insiders say AI has more than 10% chance of ending humanity – soon

The stunning admissions from three AI researchers prompted calls for urgent government action from both sides of the political divide in Washington, despite Donald Trump’s opposition.

ABC

Anthropic researcher says some developers believe AI 'could kill us all'

A researcher for AI developer Anthropic has quit the company and is one of a growing number of AI experts to call on the industry to slow down.

More Technology stories

Latest cross-verified stories

Browse all stories from September 2026 in the archive.