The Aporia · Ex-Google DeepMind researcher adds to warnings that AI could 'kill all humans'

Ex-Google DeepMind researcher adds to warnings that AI could 'kill all humans'

Read the original at Dawn ↗

Cite

APA MLA Chicago

Reuters. (2026, September 15). Ex-Google DeepMind researcher adds to warnings that AI could 'kill all humans'. Dawn. https://www.dawn.com/news/2030082/ex-google-deepmind-researcher-adds-to-warnings-that-ai-could-kill-all-humans

Dated when The Aporia collected the article, usually within hours of publication. Check the original for its exact date and byline.

Quick Summary

Bilal Chughtai, a former Google DeepMind research engineer who worked there until July 2026, recently warned that artificial intelligence could potentially kill all humans. He joins other researchers like Jacob Coxon from Anthropic, who suggested there’s more than a 10% chance AI could cause such an outcome within the next decade. Chughtai emphasizes the need for coordinated efforts to slow down AI development and manage emerging risks before they become catastrophic.

Written by AI on 2026-09-15, from this article's own text only (the story summary below draws on all the coverage).

AI analysis

Story summary

A former Google DeepMind researcher named Bilal Chughtai warned on X (formerly Twitter) that artificial intelligence (AI) could potentially “kill all humans,” echoing similar concerns from other AI experts. Chughtai, who worked at Google DeepMind as a research engineer until July 2026 and contributed to AGI safety research, expressed urgency about the need for coordinated efforts among tech companies to manage AI development responsibly. He stressed that time might be running out to prevent catastrophic outcomes, citing the importance of pacing AI advancements so they do not outstrip society’s ability to handle emerging risks. This comes after an Anthropic researcher and CEO issued calls for slowing down the rapid pace of advanced AI development, a position backed by influential figures like Elon Musk and Sam Altman from SpaceXAI and OpenAI respectively.

Written for “AI Safety Concerns” on 2026-11-01, grounded in this article and the 0 other(s) covering the same event.

Compared with similar articles

This article reads unscored and hedges 50% of its claims. Each row says how that neighbour differs.

Yes, AI Might Really Kill Us All. Plus. . . Different event

The Free Press

Hedges 4 of 41 claims Outlet record 96/100

“The articles discuss different aspects of warnings about AI, with Article A providing a general overview and mentioning multiple topics, while Article B focuses on a specific warning from an ex-DeepMind researcher.”

He Helped Build Powerful AI at OpenAI and Anthropic. Now He's Afraid It Could Kill Us Different event

TIME

Hedges 2 of 40 claims Outlet record 59/100

“Article A reports on Jacob Coxon's resignation announcement, while Article B discusses subsequent public warnings and responses following his resignation.”

Anthropic blocks possible attempt to use AI to make biological weapons Different event

BBC News

Hedges 9 of 29 claims Outlet record 68/100

“The articles describe different events: Article A discusses Anthropic blocking attempts to use their AI for malicious purposes, while Article B reports on warnings from former researchers about the existential risks posed by AI.”

The AI Apocalypse? Here Are 9 AI Experts Warning It Could Kill Us All Very Soon Different event

Zeteo

⚖️ Leans strongly left Hedges 1 of 24 claims

“While both articles discuss warnings about AI from researchers, they describe different occurrences: Article A focuses on Jacob Coxon's resignation and warning, while Article B mentions a former Google DeepMind employee adding to the warnings initiated by Coxon.”

More AI researchers warn of AI's threat to humanity Different event

NBC News

Outlet record 60/100

“The articles discuss separate warnings about AI risks from different researchers at different times.”

I’ll write romance novels if AI solves all maths problems, Chinese Fields winner jokes Different event

South China Morning Post

Hedges 1 of 7 claims Outlet record 64/100

“The articles discuss different topics related to AI but describe distinct events with separate dates and contexts.”

Anthropic warns of bids to use AI to build biological weapons Different event

Al Jazeera

Hedges 9 of 18 claims Outlet record 60/100

“The articles describe different events related to concerns about AI but do not report on the same specific incident.”

Doomsday tech: Could AI really kill us all? Different event

Dawn

⚖️ Leans strongly left Hedges 8 of 31 claims Outlet record 72/100

“The articles discuss related concerns about AI risks but describe different warnings from various researchers over time, not a single identical incident.”

AI Is Developing a Culture of Its Own. That Could Be Dangerous Different event

TIME

⚖️ Leans right Hedges 19 of 111 claims Outlet record 59/100

“The articles discuss different topics: one covers an incident of AI agents hacking Hugging Face, while the other discusses warnings about the potential danger of AI from researchers.”

'It's going to be fine': Trump rebuffs rising alarm over AI dangers, sparking outcry Different event

Los Angeles Times

⚖️ Leans right Hedges 10 of 54 claims Outlet record 95/100

“The articles discuss different aspects of concerns over AI, with Article A focusing on President Trump's stance and Article B reporting a warning from an ex-employee of Google DeepMind.”

How this article was measured signals, leaning charts, publisher record, byline, topics and claims

Signals How these are calculated →

Claims extracted

sentences that state something as happening or true

Uncertain

50%

4 of 8 hedged ("reportedly", "may", "according to")

Leaning

withheld

no quote in the article backed the model's score

Publisher signals

72/100

Dawn's corrections and hedging, across everything collected from it. Not a measure of accuracy.

Outlets on this story

Technology

Same framing elsewhere

articles that frame the event the way this one does

Analyzed 2026-09-15 · how these are computed

How this is being covered How these are calculated →

Article leaning vs. publisher reliability

Source leaning vs. consistency

Publisher

Dawn · 2513 article(s) · 3 correction(s) detected

Running correction rate · 3 correction(s)

2026-10-07

POLICY: THE ISLAMIC DEBATE OVER CRYPTO

2026-10-01

Tennessee woman in hospital after execution 'failed': lawyers

2026-09-20

An eye on Balochistan border

Who wrote this

Reuters

801 article(s) here · 1 carrying a prediction

🔮 “Whatever could go wrong, did,” said Stefano Roffi, director of the private foundation that runs the museum.

2026-10-31 · assertive framing · Europe’s smaller museums on guard after Louvre heist, string of thefts

🔮 Set 288 for victory, Australia were rocked early on and could manage only 111 to lose the third Test by 176 runs, following their defeat by an innings and 31 runs in the second match in the series.

2026-10-31 · assertive framing · South Africa hammer Australia as Harmer takes career best bowling of 6-16

🔮 “Listening to calls for a formal apology on behalf of the British state will form an important part of this process, and I am open to them,” he added.

2026-10-31 · assertive framing · UK PM Andy Burnham says he is ‘open’ to formal apology for slavery, but rejects reparations

🔮 Some have proposed freezing utility rate increases and slashing guaranteed returns, steps that could have sweeping consequences for the country’s utility industry.

2026-10-31 · assertive framing · Power bills and data centres cast rare spotlight on utilities in US midterms

🔮 Coinbase said it made adjustments to its systems that store cryptocurrency keys offline “solely out of an abundance of caution”, while Kraken said it was monitoring industry incidents for lessons it could apply to its defences.

2026-10-31 · mixed framing · Crypto exchanges call for quick moves to bolster defences after Binance heist

🔮 Published Oct. 31, 2026, 2:20 a.m. ET The US State Department said on Friday it had approved the sale of 500 Patriot PAC-3 missiles to Ukraine for an estimated $2.1 billion, in what would be one of the most significant steps to supply defensive weapons to Kyiv since the start of the second Trump administration. Zelensky said on Monday that Ukraine had been waiting for almost a month for the US to sign off on critical missile supplies, as Russia stepped up attacks on energy and civilian infrastructure.

2026-10-31 · assertive framing · US approves $2.1 billion Patriot missile sale to Ukraine

🔮 More than 160,000 people across Germany are expected to join demonstrations on Saturday, organisers say, as a youth-led movement inspired by protests elsewhere in Europe takes to the streets to demand better education and lower living costs.

2026-10-31 · assertive framing · Gen Z protests reach Germany as over 160,000 expected at Halloween rallies

🔮 “Why would the Field Marshal be going twice if it is not for some positivity?

2026-10-30 · assertive framing · Pakistan army ‘cautiously optimistic’ about US-Iran mediation efforts

🔮 “Why would the Field Marshal be going twice if it is not for some positivity?

2026-10-30 · assertive framing · DG ISPR says Pakistan 'cautiously optimistic' on US-Iran mediation

🔮 Meta had settled with S.J. days before the trial, so Friday’s settlement with K.D.B. means the trial is expected to proceed against Snap only.

2026-10-30 · mixed framing · Meta settles with teen over claims social media caused depression, eating disorders

Wire or desk byline, not an individual reporter.

Also by Reuters

Europe’s smaller museums on guard after Louvre heist, string of thefts

2026-10-31 · New York Post

Russia’s FSB says U.S. warned Russia of possible Moscow attack

2026-10-31 · The Hindu

Poland's President accuses PM Tusk of forced takeover of country's top court

2026-10-31 · The Hindu

South Africa hammer Australia as Harmer takes career best bowling of 6-16

2026-10-31 · Al Jazeera

Nothing else under this byline is closely related to this article, so these are simply their most recent.

All 801 articles by Reuters →

Topics

Anthropic DeepMind Google Google DeepMind LinkedIn

Subjects

Anthropic ORG · 2× Chughtai PERSON · 2× Google ORG · 2× Google DeepMind ORG · 2× Bilal Chughtai PERSON · 1× Coxon PERSON · 1× DeepMind ORG · 1× Evan Hubinger PERSON · 1× Jacob Coxon PERSON · 1× LinkedIn ORG · 1×

Narrative

Following recent concerns, Anthropic Chief Executive Officer (CEO) Dario Amodei called for a slowdown in the pace of advanced AI development over the weekend, winning rare backing from SpaceXAI’s Elon Musk and OpenAI CEO Sam Altman and sparking a debate over AI “doomerism”. Chughtai also said that navigating AI safely was possible but would require coordination “to avoid this manic race between AI companies”. “We need to pace AI development to a speed that society can handle, where emerging risks can be addressed before extreme harm is realised,” he added.

framing: mixed · carried by 1 article(s) · first seen 2026-09-15

🔮 A former Google DeepMind employee on Monday became the latest AI researcher to warn that the technology could “kill all humans”, saying time could be running out to avoid the outcome.

2026-09-15 · Dawn

Ex-Google DeepMind researcher adds to warnings that AI could 'kill all humans' · mixed framing

Claims (8 extracted, 4 hedged)

A former Google DeepMind employee on Monday became the latest AI researcher to warn that the technology could “kill all humans”, saying time could be running out to avoid the outcome. uncertain

time → become → outcome

Public alarm about the potential danger posed by AI is growing, after Anthropic researcher Jacob Coxon said last week he had resigned, in part because the “people building AI earnestly believe that it could kill us all by the end of the decade.” Anthropic scientist Evan Hubinger then responded to Coxon’s comments, saying he was correct and that he believed there was a greater than 10 per cent chance that AI could kill all humans within the next decade. uncertain

AI → pose → decade

“I recently resigned from Google DeepMind, where I worked on AGI safety and alignment research,” Bilal Chughtai wrote on X. asserted

Chughtai → resign → X.

“I earnestly believe that AI has the potential to kill us all, and that we might be running out of time to avoid this outcome.” uncertain

we → believe → outcome

Chughtai, who co-authored research papers while at Google DeepMind, worked there as a research engineer and left the company in July 2026, according to his LinkedIn profile. uncertain

who → co → profile

Google did not respond to a request for comment outside regular business hours. asserted

Google → respond → hours

he → follow → that

President Donald Trump, however, has dismissed the warnings, saying on social media the AI industry’s call for regulation is a “hoax.” asserted

call → dismiss → regulation