P(doom) is the probability of catastrophic or existential risk from advanced AI. It estimates the chance that AI development leads to outcomes that would permanently curtail humanity's potential or cause human extinction. The term is used colloquially in AI safety research.

What are the main AI risk scenarios?

Main scenarios: (1) Alignment failure — AI pursues mis-specified goals with harmful consequences. (2) Misuse — deliberate misuse of powerful AI for harm (bioweapons, cyberattacks). (3) Loss of control — AI systems become uncontrollable or unpredictable. (4) Systemic risks — AI causing societal collapse through economic or political disruption.

What is AI alignment?

AI alignment is the problem of ensuring AI systems pursue goals that match human values and intentions. It includes technical challenges (specifying objectives, preventing reward hacking) and governance challenges (oversight, deployment decisions). Alignment research aims to reduce P(misalign|AGI).

What is the probability of AGI?

Expert surveys show wide disagreement. Median estimates for AGI by 2050 range from 10–50%. By 2100, median estimates are 50–80%. Probability depends on assumptions about compute scaling, algorithmic progress, and hardware advances. This calculator allows you to set your own estimate.

How can AI risk be reduced?

Risk reduction strategies: (1) Technical alignment research (RLHF, interpretability, robustness). (2) Governance and regulation (international coordination, safety standards). (3) Compute governance (monitoring large training runs). (4) Defensive applications (AI for cybersecurity, threat detection). (5) Gradual deployment and testing.

What are expert estimates of P(doom)?

Expert estimates vary widely. AI safety researchers' median P(doom) estimates range from 1–10% (optimists) to 20–50% (concerned). Some experts estimate >50%. The distribution is highly skewed — a minority of experts assign very high probabilities. This calculator lets you explore your own estimate.

P(doom) Calculator – AI Doom Probability Estimator

⚠️ This is an educational and exploratory tool. These estimates are deeply uncertain. Experts range from <1% to >50% P(doom). Use this to structure your thinking, not as a definitive answer.

Risk Factor Assessment

P(Superintelligence by 2100) – How likely is AGI/ASI to be developed?

70%

P(Alignment failure | AGI) – If AGI is built, what's the chance it's misaligned with human values?

30%

P(Catastrophic outcome | Misalignment) – If misaligned AGI exists, chance it causes civilisation-level harm?

40%

P(Malicious use causing catastrophe) – Chance of deliberate catastrophic misuse (bioweapons, cyberattacks)?

10%

P(Governance succeeds) – How confident are you that international AI governance prevents worst outcomes?

30%

Your P(doom)

0%

–

Alignment path

0%

Malicious use path

0%

Expert Estimates (for reference)

Eliezer Yudkowsky (MIRI): ~99%

Geoffrey Hinton: ~10–20%

Paul Christiano (ARC): ~10–20%

Yann LeCun (Meta): <1%

Sam Altman (OpenAI): Low, manageable

2022 AI Researcher Survey median: ~5%

Paste into your website:

Free to embed. No attribution required. calcuja.com

→ P(doom) Survey 2026: What researchers estimate

// What Is a Doom Calculator?

A doom calculator estimates the probability of catastrophic outcomes from advanced AI systems. In AI safety research, this probability is called P(doom) — the likelihood that artificial general intelligence (AGI) causes permanent, large-scale harm to humanity.

Unlike most calculators, there is no single formula. P(doom) is a subjective probability that depends on your assumptions about AI development timelines, alignment difficulty, and governance effectiveness. This doom calculator walks you through each assumption and combines them into a personal estimate.

Published estimates range from near 0% (Yann LeCun) to over 90% (Eliezer Yudkowsky). The median across 50+ public estimates is approximately 20%. See the full AI Risk Survey 2026 →

// The Formula

P(doom) = 1 − (1 − Path₁) × (1 − Path₂)

Path₁ = P(AGI) × P(Misalignment | AGI) × P(Catastrophe | Misalignment) × (1 − P(Governance)) · Path₂ = P(AGI) × P(Malicious use) × (1 − P(Governance))

>99%

Yudkowsky

MIRI founder — most pessimistic major researcher

10–20%

Hinton / Christiano

Geoffrey Hinton & Paul Christiano (ARC) estimates

~5%

Median researcher

2022 AI researcher survey median P(doom)

<1%

LeCun

Yann LeCun (Meta) — most optimistic major researcher

Uncertainty Quantification

The 100× Disagreement Matters

When experts disagree by two orders of magnitude (<1% vs >99%), the takeaway is not that one side is right — it's that we face genuine uncertainty about unsolved problems. P(doom) is a tool for structuring uncertainty, not a prediction. The most rational response is to support AI safety research while acknowledging the limits of our current knowledge.

// Understanding the debate

Why Experts Disagree by 100×

P(doom) estimates range from <1% to >99% because they depend on assumptions about unsolved open problems: whether the alignment problem is tractable, how quickly AI capabilities will advance, and whether governance mechanisms can keep pace with technology.

These are not empirical disagreements resolvable by more data — they are partly philosophical judgements about the nature of intelligence, human institutions, and the behaviour of optimisation processes at scale.

Educational tool: This calculator structures your uncertainty — it does not produce a ground truth. P(doom) is not a scientific metric with a knowable true value. Use it to explore your own assumptions.

// Key factors in the debate

What Drives Estimates Up or Down

Factors that increase P(doom)

Rapid capability scaling outpacing safety, race dynamics between labs or nations, difficulty of the alignment problem, inadequate governance, and dual-use misuse potential of powerful models.

Factors that decrease P(doom)

Capability plateau before dangerous levels, alignment proving tractable, international cooperation, significant lab safety investment, democratic oversight, and narrow AI remaining dominant longer than expected.

What you can do

Engage with AI governance debates. Support responsible AI institutions (ARC, MIRI, Anthropic safety teams, Centre for AI Safety). Stay informed. The civic response is more impactful than individual behaviour changes.

// Common Questions

Frequently Asked Questions

P(doom) is the estimated probability that advanced AI development leads to catastrophic or existential outcomes for humanity. It is a heuristic used by AI safety researchers, not a formal scientific metric. Estimates range from <1% to >99%, reflecting genuine and deep disagreement about unsolved problems.

Geoffrey Hinton estimates 10–20% chance of existential risk. Eliezer Yudkowsky (MIRI) expresses P(doom) > 99%. Paul Christiano (ARC) estimates ~10–20%. Stuart Russell regards the risk as serious and motivating. Yann LeCun (Meta) considers current AI architectures incapable of posing such risks. The disagreement is real and substantive.

Rapid capability advancement leaving safety work behind, lack of alignment progress, competitive race dynamics between nations and companies, the difficulty of the alignment problem, inadequate governance, and misuse potential by malicious actors with access to powerful models.

AI capability plateaus before dangerous levels, alignment proving tractable, international cooperation developing, significant AI lab safety investment, democratic oversight mechanisms, and narrow AI remaining dominant far longer than pessimists expect.

This calculator is a thought experiment and educational tool, not actionable risk management. P(doom) estimates are a prompt to engage with AI governance debates and support safety research. Most impactful actions: engage with AI policy, support responsible AI institutions, stay informed.

The alignment problem is the challenge of ensuring that advanced AI systems pursue goals that are beneficial to humanity — even as they become more capable. It is hard because we cannot perfectly specify human values, and because capable optimisers may find unexpected ways to satisfy specified goals that violate the spirit of our intentions.

Artificial General Intelligence (AGI) is typically defined as AI that can perform any cognitive task a human can. Estimates of arrival range from 2028 (optimistic) to never (sceptics). The 2022 AI Impacts survey found median researcher estimates of ~50% chance by 2059. Timelines have shortened significantly since GPT-4.

Yes, meaningfully. Constitutional AI (Anthropic), RLHF fine-tuning, interpretability research (Mechanistic Interpretability), and scalable oversight are active areas. Whether progress is fast enough relative to capability scaling is the core dispute. Funding for safety research has grown substantially since 2022.

// Methodology & References

Formula & Calculation Method

Bayesian Decomposition of P(doom)

P(doom) = P(AGI) × P(misalign|AGI) × P(catastrophe|misalign) + P(misuse) − overlap

P(AGI) — Probability of artificial general intelligence by 2100
P(misalign|AGI) — Probability of value-misalignment given AGI
P(catastrophe|misalign) — Probability of irreversible catastrophe given misaligned AGI
P(misuse) — Probability of catastrophic deliberate misuse (bioweapons, etc.)

Source: Bayesian decomposition framework adapted from MacAskill, 'What We Owe The Future' (2022); Ord, 'The Precipice' (2020)

Authoritative Sources

Ord, T. (2020): The Precipice: Existential Risk and the Future of Humanity. Hachette Books. Estimates P(doom from unaligned AI) ≈ 10% this century. Foundational framework for existential risk quantification.
MacAskill, W. (2022): What We Owe The Future. Basic Books. Bayesian expected-value framework for long-termism and risk decomposition. → whatweowethefuture.com
Grace, K. et al. (2024): "Thousands of AI authors on the future of AI." arXiv:2401.02843. Survey of 2,778 AI researchers: median P(extremely-bad outcome) ≈ 5%, median AGI by 2047. → arXiv
AI Impacts (2022): "2022 Expert Survey on Progress in AI." 738 ML researchers. Median P(human extinction or permanent severe curtailment) from AI ≈ 5%. → AI Impacts
Metaculus (2024): Community aggregated forecasts on AGI timelines and existential risk. Median AGI by 2031 (as of 2024). → Metaculus
Bostrom, N. (2014): Superintelligence: Paths, Dangers, Strategies. Oxford University Press. Foundational analysis of orthogonality thesis and instrumental convergence as drivers of existential AI risk.
MIRI (2023): Machine Intelligence Research Institute. Eliezer Yudkowsky's published position: P(doom) > 99% under current trajectories. → intelligence.org

Expert Insights & Research

Expert P(doom) estimates span ~6 orders of magnitude: Yann LeCun (Meta) <1%; Yoshua Bengio ~20%; Geoffrey Hinton ~50%; Eliezer Yudkowsky >95%. The 2022 AI Impacts Expert Survey (738 ML researchers) found median P(extremely-bad outcome) ≈ 5%.

— AI Impacts Expert Survey 2022 / Hinton interviews (2023) / LeCun Twitter (2023)

"I think it's quite likely — like 10 to 20 percent — that we'll end up with a world that is really bad for humans. I'm not sure I believe it, but I think it's a credible scenario."

— Geoffrey Hinton ("Godfather of AI"), MIT Technology Review interview (2023)

Disagreement between experts is mostly driven by disagreement about P(misalign|AGI), not P(AGI). Most researchers agree AGI is plausible by 2050; the contention is whether alignment is solvable in time.

— Grace et al., 'Thousands of AI authors on the future of AI' (2024)

// Explore More

Related Tools

AI

AI Job Displacement Risk

Estimate your role's AI automation risk and financial runway.

Calculate →

🌐

Plan B Readiness Calculator

Assess diversification needs based on political and geopolitical risks.

Calculate →

💰

Billionaire Time Calculator

How long to earn a billionaire's wealth at your salary?

Calculate →

$

Investment Calculator

Compound interest and growth projections for your investments.

Calculate →

🛡

Doomsday Prep

Calculate your emergency preparedness budget.

Calculate →

∞

Longevity Suite

Biological age and healthspan calculators.

Explore →