Evidence quality 4.88/5
Eight-dimension review score against the quality rubric . Each dimension scored 1–5.
- D1 Source grounding
- 5/5
- D2 Source authority
- 5/5
- D3 Arithmetic
- 5/5
- D4 Uncertainty
- 5/5
- D5 Scope
- 5/5
- D6 Prose
- 5/5
- D7 Perception honesty
- 4/5
- D8 Caveat completeness
- 5/5
How the risk varies
The headline figure averages across very different situations. Here’s how the probability varies by scenario or context:
—
Grace et al. 2024: median 5%, mean 9%, conditional on achieving human-level AI. Between 38-51% of respondents gave ≥10%.
—
10% unconditional over the next century, from The Precipice (2020). Largest single contributor to his overall 1-in-6 existential risk estimate.
—
Hinton: 10-20% chance of extinction within 30 years. Bengio: ~20% for AI existential risk. Both shortened their timelines post-GPT-4.
—
Effectively negligible on relevant timelines. Argue current AI paradigm cannot produce autonomous goal-seeking agents. LeCun calls x-risk debate 'complete B.S.'
—
Bimodal: 77% 'concerned' about AI risk to humanity (YouGov 2024), but only 4% name AI as most likely cause of extinction (Rethink Priorities). Concern conflates job loss, surveillance, and existential scenarios.
Bar length and shade rank these scenarios against each other, not against other risks. The exact odds are shown beside each.
The most commonly cited number in AI existential risk discourse comes from Grace et al.’s 2024 survey of 2,778 researchers who had published in top-tier AI venues: a median 5% probability (mean 9%) of human extinction or equivalent catastrophe, conditional on the development of human-level machine intelligence. Between 38% and 51% of respondents assigned at least a 10% chance. Separately, philosopher Toby Ord estimated a 10% unconditional probability of existential catastrophe from unaligned AI this century — the single largest contributor to his overall 1-in-6 estimate of existential risk from all sources. Geoffrey Hinton puts the figure at 10-20% within thirty years; Yoshua Bengio at roughly 20%. On the other side, Yann LeCun calls the entire debate “complete B.S.,” arguing that current architectures lack the persistent memory, causal reasoning, and physical-world models that autonomous goal-pursuit would require. The range of expert opinion — from approximately zero to one in five — is itself the most important datum.
The perception gap here is unusual. For most entries on this site, the public overestimates a well-measured risk. For AGI existential risk, there is no well-measured risk to over- or underestimate. Public opinion is bimodal: a 2024 YouGov survey found 77% of Americans “concerned” about AI as a risk to humanity, yet a Rethink Priorities survey found only 4% named AI as the most likely cause of human extinction — nuclear war led at 42%. Pew Research reports that 57% of Americans rate AI’s societal risks as “high,” but this figure conflates job displacement, misinformation, and surveillance with extinction scenarios. The expert community, meanwhile, clusters around a surprisingly non-trivial median — 5% is not a number most people would accept for any other cause of extinction without demanding extraordinary evidence. The fact that thousands of researchers who build these systems assign that probability is either a damning indictment of the technology’s trajectory or a reflection of how poorly calibrated even domain experts are when forecasting unprecedented events.
This entry carries a no_reliable_estimate flag because there is no defensible way to
compress the available evidence into a single lifetime probability. The event has no base
rate. The technology described by the risk scenarios does not yet exist. The expert estimates
are opinions — informed, considered, published in peer-reviewed venues, but opinions
nonetheless. The Grace et al. survey has been criticized for framing effects; the respondents
are self-selected from those willing to speculate about low-probability catastrophes; and the
conditional nature of the question (extinction given human-level AI) leaves the
unconditional probability dependent on a timeline forecast that itself spans decades. What
can be said honestly: a non-trivial fraction of the people who understand AI best think there
is a non-trivial chance it ends very badly, and a similarly credentialed fraction thinks the
concern is premature to the point of absurdity. Both positions are held by Turing Award
laureates. The honest answer is that nobody knows, and anyone who tells you otherwise is
selling something — possibly an AI safety startup, possibly a GPU.
Related tidbits
Expert median estimate for catastrophic AGI risk: ~5%. Nuclear weapons have gone unused in conflict for 80 years despite ~13,000 warheads. One risk is theoretical. The other has a track record of near-misses.
Claim ledger
Every number below is what each source reported, with the verbatim quote we relied on and how we arrived at our figure. Click any link to verify directly.
2/6 sources independently verified verbatim against the cited source
-
[1] arXiv (Grace, Stewart, Sandbrink, Thomas, Weinstein-Raun, Brauner) — Thousands of AI Authors on the Future of AI Verified
Thousands of AI Authors on the Future of AI- Statistic
Median 5% probability of human extinction or similarly permanent catastrophe from advanced AI; mean 9%; 38-51% of respondents assigned ≥10% probability- Excerpt
“"Between 38% and 51% of respondents gave at least a 10% chance to advanced AI leading to outcomes as bad as human extinction." ”
- Source data from
- 2024-01-05
- Accessed
- 2026-04-19 · archived copy
- Verification
- Excerpt independently re-fetched and confirmed word-for-word against the cited source during our grounding audit.
- Calculation
- Grace et al. surveyed 2,778 researchers who had published in top-tier AI venues (NeurIPS, ICML, ICLR, AAAI, IJCAI, JMLR). The survey asked for probability estimates of "extremely bad" outcomes (human extinction or equivalent) conditional on the development of human-level machine intelligence. The median response was 5%, the mean 9%. Between 38% and 51% of respondents — depending on question framing — assigned at least a 10% probability. No conversion to an unconditional lifetime probability is attempted here because the conditional (human-level AI existing) is itself deeply uncertain, and multiplying two poorly constrained quantities yields a number that conveys false precision.
- Independence
- Largest survey of AI researchers to date. Methodologically independent of Ord's philosophical risk analysis and of public polling. The survey population overlaps with the AI safety research community but also includes many researchers with no prior engagement with existential risk discourse.
-
[2] Bloomsbury Publishing (Toby Ord, Future of Humanity Institute, Oxford) — The Precipice: Existential Risk and the Future of Humanity
The Precipice: Existential Risk and the Future of Humanity- Statistic
1 in 10 (10%) existential risk from unaligned AI this century — the largest single contributor to Ord's overall 1-in-6 existential risk estimate- Excerpt
“"I put the existential risk from unaligned artificial intelligence at around 1 in 10 over the next hundred years. This is higher than all other sources of existential risk I consider, combined." ”
- Source data from
- 2020-03-06
- Accessed
- 2026-04-19 · archived copy
- Calculation
- Ord's 10% figure is a subjective probability estimate informed by consultation with AI researchers, historical precedent in technology governance, and analysis of alignment failure modes. It is explicitly not an empirical measurement — it is the considered judgment of a moral philosopher specializing in existential risk. Ord estimates total existential risk this century at ~1 in 6 (17%), with AI contributing 10 percentage points, engineered pandemics ~3%, and other sources (nuclear, climate, asteroid, supervolcano) contributing the remainder. The 10% figure is unconditional on AGI development — it bakes in Ord's timeline uncertainty about when or whether such systems will be built.
- Independence
- Ord's framework is a philosophical risk analysis drawing on the broader Future of Humanity Institute research program. Independent methodology from the Grace et al. survey, though Ord cites earlier versions of the same survey (2016 edition) as one input among many.
-
[3] Pew Research Center — How the US Public and AI Experts View Artificial Intelligence
How the US Public and AI Experts View Artificial Intelligence- Statistic
50% of Americans are more concerned than excited about AI's increased role in daily life (up from 37% in 2021); 57% rate the risks to society as 'high'- Excerpt
“"A majority of Americans (57%) rate the risks of AI for society as high, with far fewer seeing high benefits." ”
- Source data from
- 2025-04-03
- Accessed
- 2026-04-19 · archived copy
- Calculation
- Pew's 2025 survey tracks general AI concern, not existential risk specifically. The 57% "high risk" figure conflates near-term harms (job loss, misinformation, surveillance) with extinction-level scenarios. A separate YouGov survey (December 2024) found 77% "concerned" about AI as a "risk to humanity," with 39% "very concerned." These figures are included to document the public perception landscape, but they cannot be compared directly to the Grace et al. conditional extinction probability because they measure different constructs. The bimodal pattern — high general concern, low salience of extinction specifically (only 4% name AI as the most likely cause of extinction per Rethink Priorities) — is itself the key finding for the perception-gap framing.
- Independence
- Pew Research Center is an independent nonpartisan research organization. Its methodology (nationally representative probability sampling) is fully independent of the AI researcher survey and of Ord's philosophical analysis.
-
[4] TechCrunch — Meta's Yann LeCun says worries about AI's existential threat are 'complete B.S.'
Meta's Yann LeCun says worries about AI's existential threat are 'complete B.S.'- Statistic
LeCun argues current AI systems lack key capabilities (persistent memory, reasoning, planning, physical-world understanding) and dismisses AI's existential threat as 'complete B.S.'- Excerpt
“"You're going to have to pardon my French, but that's complete B.S." LeCun argued that today's large language models lack some key cat-level capabilities, like persistent memory, reasoning, planning, and an understanding of the physical world. ”
- Source data from
- 2024-10-12
- Accessed
- 2026-04-19 · archived copy
- Calculation
- LeCun (Turing Award 2018, Chief AI Scientist at Meta) and Andrew Ng (Stanford, Coursera co-founder) represent the most prominent skeptical position among credentialed AI researchers. Their argument is architectural: current LLMs "manipulate language" without genuine understanding, lack persistent memory and causal reasoning, and are fundamentally incapable of the autonomous goal-pursuit that x-risk scenarios require. LeCun's blanket dismissal of the existential-threat worry as "complete B.S." is contested — the Grace et al. survey shows a median 5% extinction probability, which is not "zero." The disagreement is partly empirical (what current systems can do) and partly about extrapolation (whether scaling or architectural breakthroughs will produce qualitatively different capabilities). No probability estimate is extracted from the skeptical position because it is framed as a categorical rejection of the premise rather than a quantified risk.
- Independence
- News reporting of LeCun's public statements. Represents a distinct intellectual position from both the Grace et al. survey respondents (who gave non-zero medians) and Ord's philosophical analysis.
-
[5] Forbes, relaying a December 2024 Guardian report on Hinton's BBC Radio 4 Today programme interview — Geoffrey Hinton's Prediction Of Human Extinction At The Hands Of AI
Geoffrey Hinton's Prediction Of Human Extinction At The Hands Of AI- Statistic
Geoffrey Hinton (2018 Turing Award; 2024 Nobel Prize in Physics) raised his estimate of AI-caused human extinction within 30 years to a 10-20% range, up from an earlier standalone 10% estimate- Excerpt
“"Geoffrey Hinton, often referred to as one of the 'godfathers' of artificial intelligence, believes there is a 10% to 20% chance that AI could drive humanity to extinction within 30 years." Speaking on BBC Radio 4, Hinton said: "we've never had to deal with things more intelligent than ourselves before." ”
- Source data from
- 2024-12-29
- Accessed
- 2026-07-01 · archived copy
- Calculation
- Hinton gave a standalone 10% catastrophic-outcome estimate in earlier 2024 interviews; in a December 2024 BBC Radio 4 interview (first reported by The Guardian, relayed here via Forbes) he raised it to a 10-20% range for AI causing human extinction within 30 years. Cited solely to source the regional_breakdown "Hinton & Bengio estimates" row and the body-text expert-range claim; the entry's no_reliable_estimate framing continues to rest on Grace et al. and Ord, not on this single-individual estimate.
- Independence
- News relay of a broadcast interview — a single named individual's stated view, methodologically comparable to the LeCun source above on the skeptical side rather than a systematic survey or philosophical analysis.
-
[6] arXiv preprint — An International Agreement to Prevent the Premature Creation of Artificial Superintelligence Verified
An International Agreement to Prevent the Premature Creation of Artificial Superintelligence- Statistic
Documents Yoshua Bengio's publicly stated estimate of roughly a 20% probability that advanced AI development turns out catastrophic- Excerpt
“"Yoshua Bengio, a Turing Award recipient and the most cited living scientist, said he thinks there is a 'twenty percent probability that it turns out catastrophic'" ”
- Source data from
- 2025-11-01
- Accessed
- 2026-07-01 · archived copy
- Verification
- Excerpt independently re-fetched and confirmed word-for-word against the cited source during our grounding audit.
- Calculation
- Bengio has not published one definitive percentage in his own formal writing — his blog FAQ on catastrophic AI risks instead proposes a compound formula whose component probabilities he says should be set by expert polling, and his recorded talks describe capability timelines rather than a single risk probability. The ~20% figure attributed to him is nonetheless widely and consistently reported; this arXiv paper is cited because it documents the specific figure in a dated, citable context. Used solely to source the regional_breakdown "Hinton & Bengio estimates" row and the body-text expert-range claim.
- Independence
- Secondary academic documentation of a public statement rather than a primary interview transcript or Bengio's own published writing. Direct checks of Bengio's own blog, a TED talk transcript, and an OECD.AI incident summary did not turn up a first-party statement of this exact figure, though the ~20% number recurs consistently across independent secondary sources.