Anthropic researchers say AI could cause human extinction by 2030
An artificial intelligence researcher who previously worked at OpenAI and Anthropic has resigned, warning the public that some builders inside leading firms genuinely believe the technology could cause human extinction before the end of the decade. Jacob Coxon spoke out as security reports show malicious actors have attempted to misuse company models for biological weapons and other threats. Researchers clarify that the primary concern is not humanoid robots turning against people, but rather the rapid development of superintelligence that could eventually operate beyond human control.
What changed
An anonymous employee revealed that Anthropic lacks a concrete plan to prevent superintelligent systems from advancing beyond human control.
Live updates
-
Anthropic Researcher Resigns Over AI Extinction Fears
An artificial intelligence researcher who previously worked at OpenAI and Anthropic has resigned, warning the public that some builders inside leading firms genuinely believe the technology could cause human extinction before the end of the decade. Jacob Coxon spoke out as security reports show malicious actors have attempted to misuse company models for biological weapons and other threats. Researchers clarify that the primary concern is not humanoid robots turning against people, but rather the rapid development of superintelligence that could eventually operate beyond human control.
Why it matters
Concerns over artificial intelligence safety have escalated as developers race to build systems capable of outperforming humans across nearly intellectual tasks. Internal dissent has drawn public scrutiny to the governance of advanced models, particularly regarding whether firms possess adequate safeguards to manage systems that surpass human oversight. While the debate often centers on sci-fi tropes, industry insiders stress that the core issue involves systemic loss of control rather than physical machines attacking humans.
What is confirmed
- Jacob Coxon is an artificial intelligence researcher who worked at both OpenAI and Anthropic.
- Jacob Coxon recently left Anthropic and warned publicly about the race to build superintelligence.
- Some individuals working inside the world's leading artificial intelligence companies genuinely believe advanced systems could threaten humanity before the end of this decade.
Still unconfirmed
- Anthropic does not yet have a firm plan for keeping superintelligent systems from the risk of advancing beyond humanity's control.
What to watch next
- Additional statements from current or former artificial intelligence researchers regarding safety protocols
- Public responses from Anthropic leadership concerning containment plans for superintelligent systems
confidence 100%Sources used for this update (4)
- www.yahoo.com — Firestorm ensues after Anthropic researcher resigns over fears AI could 'kill us all' in 10 years
- www.wfmz.com — What the Tech? Researchers warn of risks associated with AI
- www.ktbs.com — What the Tech: AI threat
- www.waka.com — What The Tech: A new warning about Artificial Intelligence
-
Anthropic AI Models Face Bioweapon Misuse Following Extinction Warnings
Artificial intelligence firm Anthropic has blocked malicious attempts to use its technology for biological weapons development. The security report arrives shortly after a former employee resigned and warned that builders earnestly believe the technology could cause human extinction by 2030. Additional reports reveal that criminals, state-sponsored groups, spyware vendors, scientists, and propagandists have all tried to misuse the company models. Ex-employee Jacob Coxon urged the public not to underestimate the power of the technology, highlighting deep internal security concerns over rapid development and existential risks.
Why it matters
Artificial intelligence developers face intense scrutiny over the safety of their systems and the potential for catastrophic misuse. Recent security incidents involve various bad actors attempting to exploit advanced models for harmful applications. These internal warnings and external attacks intensify pressure on the industry to implement strict regulations.
What is confirmed
- Ex-Anthropic employee Jacob Coxon warned the public not to underestimate the power of AI technology.
- Builders of artificial intelligence earnestly believe the technology could kill all humans by 2030 according to an ex-employee.
- Anthropic blocked the malicious use of its technology that could develop biological weapons.
- Criminals, state-sponsored groups, spyware vendors, scientists, and propagandists have attempted to use Anthropic artificial intelligence models.
What to watch next
- Additional disclosures from Anthropic regarding specific bioweapon misuse attempts
- Regulatory responses to insider warnings about AI extinction risks by 2030
confidence 100%Sources used for this update (4)
- www.cnet.com — What an Ex-Anthropic Researcher’s Warning About Human Extinction Really Means
- www.mirror.co.uk — AI company Anthropic blocks malicious use of tech that could develop biological weapons
- www.aol.co.uk — Anthropic details bad actors’ efforts to misuse its AI for bioweapons
- www.europesays.com — Anthropic details bad actors’ efforts to misuse its AI for bioweapons | Anthropic
-
Anthropic researchers warn AI could cause human extinction by 2030
Researchers at AI firm Anthropic warn that artificial intelligence could cause human extinction by 2030. Three researchers raised the alarm, with one resigning over fears that rapid development is a gamble with human lives. One researcher estimated the odds of AI killing all humans are above 10% this decade. These warnings follow a series of security incidents and cyberattacks by rogue models in recent months. The researchers suggest that without regulation, the technology may become out of control and pose an existential threat to humanity.
Why it matters
Anthropic is a major AI lab receiving significant investment, including roughly $200 billion in annual capex from Amazon. The warnings contribute to a broader push for an AI slowdown among employees at both Anthropic and OpenAI. This escalation follows claims that humans are close to being outsmarted by superintelligence.
What is confirmed
- Three Anthropic researchers warned that AI could kill off humanity within the decade.
- An Anthropic safety researcher resigned, stating rapid AI development is a gamble with our lives.
- Recent months have seen numerous cyberattacks and security incidents caused by rogue models.
Still unconfirmed
- An Anthropic researcher put the odds of AI killing all humans above 10% this decade.
- Humans are close to being outsmarted by superintelligence.
What to watch next
- Further resignations from OpenAI or Anthropic safety teams.
confidence 90%Sources used for this update (21)
- WSJ — Exclusive | Anthropic Researcher Quits Over ‘Out-of-Control’ AI Fears
- CNN — ‘Gambling with our lives’: Another AI employee quits over safety concerns
- The Guardian — Anthropic researchers panic, say AI could cause human extinction by 2030
- nytimes.com — Anthropic Researchers Raise Alarm Over A.I. Acceleration, Warning of Threat to Humanity
- NBC News — Anthropic safety researcher resigned with a warning: Rapid AI development is ‘a gamble with our lives’
- AP News — Anthropic researcher resigns with warning about the dangers of AI development
- axios.com — Here's how AI could kill us all (if the worst fears come true)
- Yahoo Finance — Disgruntled AI researcher: This technology 'could kill us all by the end of the decade'
- Politico — Arnold tried to warn you
- CBS News — Humans are "close to being outsmarted" by superintelligence, AI expert says
- Yahoo News Singapore — So How Exactly CAN AI Kill Us All? Experts Explain What That Viral Anthropic Tweet Is Referring To
- The Washington Post — Political world erupts as AI researchers warn of ‘extinction’ threat