Live verdicts on Aschenbrenner's Situational Awareness predictions — AGI timelines, graded against reality. https://agiscorecard.com/ 2026-07-26T00:00:00Z The AGI Scorecard CC BY 4.0 — reuse with attribution and a link to https://agiscorecard.com https://agiscorecard.com/for-agents 2026-07-26T00:00:00Z

Free, machine-readable AGI verdict data: /data.json (CC BY 4.0), an Atom feed, and llms.txt. Copy-paste prompt to add verdict-change alerts to any AI agent.

https://agiscorecard.com/progress-index 2026-07-26T00:00:00Z

A single auditable 0-100 score of how much of Aschenbrenner's Situational Awareness thesis is holding up. Currently 62.5/100, from 8 graded verdicts. CC BY 4.0.

https://agiscorecard.com/forecaster-leaderboard 2026-07-26T00:00:00Z

Who's winning the AGI bet? Musk says 2026, Aschenbrenner 2027, most experts ~2030 — a ranked board of every public AGI forecaster vs the date they called.

https://agiscorecard.com/ai-tools 2026-07-26T00:00:00Z

Six free AI tools, no sign-up: find your AGI type, check your job's AI exposure, bet on 12 predictions, and read the live 0-100 AGI-2027 score. Three are embeddable.

https://agiscorecard.com/prediction-receipts 2026-07-26T00:00:00Z

Musk said AGI by end of 2026. Amodei 2026–27. Aschenbrenner 2027. Live countdowns to every dated AGI prediction — the receipts stay when the dates pass.

https://agiscorecard.com/elon-musk-agi-prediction 2026-07-25T00:00:00Z

Elon Musk (xAI) has suggested AGI by end of 2026 — the most aggressive timeline among prominent figures, ahead of Aschenbrenner's 2027.

https://agiscorecard.com/cn 2026-07-25T00:00:00Z

他写下《态势感知》预言 AGI 2027，靠这套逻辑两年回报超10倍、管理200亿美元。但他最新持仓显示：一边重仓 AI 基建，一边用85亿美元看跌期权做空英伟达等芯片股。他在赌什么？

https://agiscorecard.com/invest 2026-07-25T00:00:00Z

From AGI predictions to AI investing: Aschenbrenner's $20B bet, how Buffett, Cathie Wood and 6 more legends are positioned on AI (Q1 2026 13F), and free tools.

https://agiscorecard.com/what-jobs-are-safe-from-ai 2026-07-24T00:00:00Z

The safest work needs reliable, unsupervised, end-to-end judgment — the exact thing AI still lacks. Why 'safe' is about autonomy, not intelligence, and how to watch it change.

https://agiscorecard.com/ai-job-risk-check 2026-07-24T00:00:00Z

Answer 6 questions about your job and get your AI exposure tier — grounded in real 2026 evidence (~83% GDPval, ~80% SWE-Bench) and the autonomy gap. No sign-up.

https://agiscorecard.com/how-close-is-agi 2026-07-22T00:00:00Z

Closer than skeptics say, further than optimists claim. Mid-2026: frontier AI clears many knowledge-work bars (~83% GDPval.

https://agiscorecard.com/who-is-leopold-aschenbrenner 2026-07-21T00:00:00Z

Leopold Aschenbrenner is the former OpenAI researcher who predicted AGI by 2027 in Situational Awareness (2024).

https://agiscorecard.com/ai-vs-human-intelligence 2026-07-21T00:00:00Z

Superhuman in narrow bands, subhuman where it counts — reliable autonomous judgment. Intelligence is a jagged frontier, not one scale. Where AI leads and humans still win.

https://agiscorecard.com/narrow-vs-general-ai 2026-07-21T00:00:00Z

Everything shipping in 2026 is still narrow AI — very wide narrow AI, but narrow. The ANI → AGI → ASI ladder, the line that separates narrow from general, and where we are.

https://agiscorecard.com/advertise 2026-07-21T00:00:00Z

Sponsor the AGI Scorecard: a weekly briefing and 97-page site reaching a high-intent audience researching AGI timelines. Introductory rates for early sponsors.

https://agiscorecard.com/future-bet 2026-07-21T00:00:00Z

Bet YES/NO on 12 bold predictions — AGI, robots, Mars, fusion, alien life — and see which real forecaster your bets match. Shareable grid, no sign-up.

https://agiscorecard.com/what-is-unhobbling-ai 2026-07-19T00:00:00Z

'Unhobbling' is Aschenbrenner's term for gains from removing artificial limits on models — tools, memory, agents, reasoning — versus raw scale.

https://agiscorecard.com/ai-2027-scenario-explained 2026-07-19T00:00:00Z

'AI 2027' is a widely-discussed 2025 scenario depicting, month by month, how a fast takeoff could unfold around 2027 via AI automating AI research.

https://agiscorecard.com/gdpval-explained 2026-07-19T00:00:00Z

GDPval is a benchmark that scores AI on real, economically valuable knowledge-work tasks across many occupations. Top models now reach ~83% — the evidence behind the On-track 'AI outpaces knowledge workers' verdict.

https://agiscorecard.com/ai-progress-2026-so-far 2026-07-19T00:00:00Z

Mid-2026 synthesis of all 8 tracked predictions: capability and compute on track, capex exceeded, the open-source call wrong, AGI-by-2027 still open.