Alex Heath alleges coordinated funding behind AI extinction warnings

The claim follows public statements from industry researchers estimating a 10 percent probability that advanced systems could wipe out humanity.

Alex Heath asserted that a coordinated, well-funded operation is behind rhetoric claiming there is a 10% chance artificial intelligence could kill everyone.¹ The allegation lands during a sharp escalation in public debate over existential risk, where several prominent researchers and technical specialists have argued that runaway machine capability poses an immediate threat to human survival.

Disclosures and internal departures

The warnings intensified after Evan Hubinger, an alignment researcher at Anthropic, wrote on social platform X that he saw a greater than 10% chance that rapid technological progress could kill all humans within the next decade.²³ Hubinger stated that Anthropic is trying its best, but noted that the company does not yet have a plan to solve alignment for superintelligence and is not clearly on track to do so.⁴³ His comments came in response to Jacob Coxon, an artificial intelligence researcher who recently resigned from Anthropic after earlier working at OpenAI.²³ Coxon wrote on X that neither company is acting responsibly in handling frontier systems, warning that technical laboratories will soon produce superhuman tools capable of hacking digital networks, altering scientific fields overnight, and acquiring independent resources.³

Geoffrey Hinton, an emeritus professor at the University of Toronto who received the 2024 Nobel Prize in Physics for foundational work on machine learning, endorsed those warnings, saying a 10% risk of human extinction seems a not unreasonable estimate.² Hinton previously quit his position as a Google vice president in May 2023 to speak freely about systemic danger.² He had earlier estimated the odds of artificial intelligence wiping out humanity at 10% to 20% over the next three decades, comparing future human supervision of machine intelligence to a three-year-old child attempting to manage an adult.²⁵ Paul Christiano, a prominent researcher who co-authored an influential 2016 safety paper with Anthropic co-founder Dario Amodei, also issued a public warning.² Christiano had long argued that advanced artificial intelligence would develop at a manageable pace and could be guided safely, but announced that he now sees meaningful risk of catastrophic and irreversible loss of control in the very near term.

The origins of the ten percent figure

The recurring 10% figure has also faced technical scrutiny from researchers examining how such estimates entered public circulation. Melanie Mitchell examined the widespread claim that half of artificial intelligence researchers believe there is a 10% chance the technology will kill humanity, tracing the cited statistic to the 2022 Expert Survey on Progress in AI conducted by AI Impacts.⁶ The survey, funded by the grant-giving arm of the defunct cryptocurrency exchange FTX, contacted approximately 4,271 people who had published papers at two machine learning conferences, NeurIPS and ICML, in 2021.⁶ Organizers received responses from 738 participants, but the specific question asking what probability respondents put on human inability to control future systems causing extinction received only 162 replies.⁶ Among those 162 respondents, the median estimated probability was 10%, meaning the widely cited consensus rested on 81 individuals giving that figure or higher.⁶

Mitchell noted that the survey question remained vague by omitting any time horizon, leaving open whether respondents were forecasting across 20 years, a century, or longer periods.⁶ She also pointed to potential response bias among authors choosing to answer the prompt.⁶ At the Machine Intelligence Research Institute, a research document authored by william outlined an even more severe assessment, stating that human extinction is the default outcome without aggressive government intervention and placing the probability above 90%. Longtermist research summaries have tracked similar divergence among specialists, contrasting researcher estimates with lower median figures from generalist forecasters.

Commercial laboratories have simultaneously reported technical strain in evaluating safety boundaries. Anthropic noted in an internal safety evaluation that it had grown less confident in its benign assessment of model risks, citing early indications that capability may be accelerating beyond original timetables. Meanwhile, political figures have sought multilateral oversight. Darren Jones, the former chief secretary to the Treasury in Britain, addressed an open letter urging the negotiation of an international treaty on advanced computing, arguing that global standards must be established before technical progress outpaces public oversight mechanisms.³

What this rests on

16 sentences trace to 6 sources.

All 6Documents 1Records 0Reporting 5People 0

1. 1 H https://x.com/alexeheath Document · twitter.com · 20 Sep 2026 Alex Heath Alleges Coordinated Funding Behind AI Extinction-Risk Rhetoric See the source
2. 2 P Primary Third-party reporting · theprimary.com · 10 Sep 2026 Geoffrey Hinton backs Anthropic researcher on 10% extinction risk See the source
3. 3 BN BBC News Third-party reporting · bbc.com · 9 Sep 2026 Anthropic researcher believes more than 10% chance AI 'could kill all humans' See the source
4. 4 A aiweekly.co Third-party reporting · aiweekly.co · 9 Sep 2026 Anthropic Alignment Lead Puts AI Extinction Odds Over 10% | AI Weekly See the source
5. 5 L longtermwiki.com Third-party reporting · longtermwiki.com The Case For AI Existential Risk See the source
6. 6 AA AI: A Guide for Thinking Humans Third-party reporting · aiguide.substack.com · 23 Apr 2023 Do half of AI researchers believe that there's a 10% chance AI will kill us all? See the source

Also consulted: 1 page across 1 site, none cited.