Live verdicts on Aschenbrenner's Situational Awareness predictions — AGI timelines, graded against reality. https://agiscorecard.com/ 2026-08-29T00:00:00Z The AGI Scorecard CC BY 4.0 — reuse with attribution and a link to https://agiscorecard.com https://agiscorecard.com/for-agents 2026-08-29T00:00:00Z

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/will-agi-arrive-2027 2026-08-29T00:00:00Z

Aschenbrenner predicted AGI by 2027. As of mid-2026 it's graded Open — 18 months left. How his timeline compares to Metaculus (2033), Hassabis (2030).

https://agiscorecard.com/situational-awareness-predictions 2026-08-29T00:00:00Z

Every major prediction from Aschenbrenner's Situational Awareness essay, tracked against reality: AGI 2027, compute scaling, open source, the Project, and more.

https://agiscorecard.com/when-will-agi-arrive 2026-08-29T00:00:00Z

Public AGI forecasts now cluster between 2026 and 2047: Musk 2026, Aschenbrenner 2027, Hassabis ~2030, Metaculus 2033, academic surveys 2047.

https://agiscorecard.com/ai-2027-scenario-explained 2026-08-29T00:00:00Z

'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/how-close-is-agi 2026-08-29T00:00:00Z

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

https://agiscorecard.com/ai-progress-2026-so-far 2026-08-29T00:00:00Z

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.

https://agiscorecard.com/sam-altman-agi-prediction 2026-08-29T00:00:00Z

Sam Altman says OpenAI is confident it knows how to build AGI, with superintelligence 'a few thousand days' away. His position vs the other forecasts.

https://agiscorecard.com/dario-amodei-agi-prediction 2026-08-29T00:00:00Z

Anthropic's CEO avoids the term AGI but says 'powerful AI' — a country of geniuses in a datacenter — could arrive as early as 2026–27.

https://agiscorecard.com/is-nvidia-overvalued 2026-08-29T00:00:00Z

We don't grade the price — we grade the bet: 90% of Nvidia's AI story rides on compute scaling and the capex wave, under 10% on AGI itself. Mapped, dated, sourced.

https://agiscorecard.com/does-copying-13f-work 2026-08-29T00:00:00Z

Backtested at filing-day closes over 8 rebalances: Druckenmiller's AI book +187.2% vs QQQ +59.6%; Buffett's AI slice lagged. Copying works — if you pick who.

https://agiscorecard.com/intelligence-explosion-2027 2026-08-27T00:00:00Z

Aschenbrenner predicted an intelligence explosion — AI automating AI research — beginning 2027–29.

https://agiscorecard.com/what-is-agi 2026-08-26T00:00:00Z

AGI is AI that can do essentially any cognitive task a skilled human can. What the term means, why definitions split the timelines, and where 2026 stands.

https://agiscorecard.com/who-is-leopold-aschenbrenner 2026-08-25T00:00:00Z

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

https://agiscorecard.com/ai-capex-trillion-dollar 2026-08-25T00:00:00Z

Aschenbrenner's most clearly vindicated call: AI infrastructure capex has run ahead of even his aggressive trillion-dollar projections.

https://agiscorecard.com/agi-odds-vs-evidence 2026-08-25T00:00:00Z

Polymarket's AGI-by-2027 contract vs the evidence: the market prices an announcement, the scorecard grades capability (62.5/100). Issue #1 of a running comparison.

https://agiscorecard.com/do-ai-trading-agents-work 2026-08-24T00:00:00Z

AI trading agents, audited: day-trading base rates, backtest overfitting, funding hype vs verifiable returns. Dated sources; not investment advice.

https://agiscorecard.com/is-claude-getting-dumber 2026-08-24T00:00:00Z

Users were right once: Anthropic's 2025 postmortem confirmed three real bugs behind 'Claude got worse'. The 2026 complaint cycle, audited — dated sources.

https://agiscorecard.com/justin-sun-2013-bitcoin-story 2026-08-23T00:00:00Z

A two-ledger audit: every number in the famous 2013 Bitcoin origin story traces only to Sun himself. What IS verified — positioning, credentials, a $1M round announced as eight figures.

https://agiscorecard.com/what-is-b-ai 2026-08-23T00:00:00Z

B.AI resells Claude, GPT and Gemini access behind one crypto-settled API key, with x402 agent payments built in. What's verified, what's self-reported, and what it signals — dated sources.