Today › Daily report
Daily reading · 30 October 2026
Oct 30: Sam Altman says OpenAI holds discoveries beyond its 700 published proofs; held-back results from a capability OpenAI has already shown move none of our numbers.
1 · Is AGI here?
Not in public: 0 of 8 parts met (2 partial, 6 far).
AGI anywhere, public or hidden: 0.9% today
- Breadth, status Far, step 1 of 4. On real freelance projects the best system delivers 21.25%, against 80%; results by occupation for the other half are unpublished.
- Quality, status Far, step 1 of 4. Wins or ties professionals in 84.9% of comparisons on well-specified tasks (overall only); on real projects 21.25%, against 50%.
- Reliability, status Far, step 1 of 4. On a hard agent test the best model gets 35.1% of tasks right on all four tries, against 80%.
- Reasoning, status Partial, step 2 of 4. Expert exams are passed (GPQA Diamond 96%); new research results by AI alone: 4 of the 10 we ask for.
- Long projects, status Far, step 1 of 4. The best measured 80% time horizon is about 3.1 hours, some 54 times short of a work-month.
- Autonomy, status Far, step 1 of 4. Hands-off on few-minute tasks (86%); on 1.6-hour workflows 44.3% strict, against 80%.
- Learning, status Far, step 1 of 4. Captures about a quarter of the available gain from experience (25.4%, against 50%); keeping it is unmeasured.
- Generalization, status Partial, step 2 of 4. At first contact with a new kind of game the best score was 1.5%, against 80%; 62.7% came only after 5 months.
1 new result pending our weekly check: reasoning.
2 · Could it be hidden?
2.5% Hidden AGI Index, no change
Above AGI anywhere because 1.65 points are hidden self-improvement (B), which needs no AGI.
3 · Should I worry today? Separate: set by published rules, never by the odds aboveRules, not odds
Fire alarm: Level 1 · Watch since Sep 29 · W3, W4, W5 met
Quiet day
Altman says OpenAI holds unpublished discoveries
Asked 'We have more than the 700 proofs?', Sam Altman answered 'We do. Of course.' and said OpenAI's math spending 'includes much more other discovery than what we have published'. He named no date, field, model or count; OpenAI is 'working through' a way for the world to benefit while people in those fields 'get the best version of the future' (Altman, via Cheeky Pint's published transcript; verified fact (part our inference)). It bears on Hidden AGI's kept-quiet question but measures none of the parts holding it down, such as long projects (our inference). verified fact (part our inference)Sam Altman, via Cheeky Pint's published transcript
Probabilities
| Hypothesis | Today | By end-2030 | By end-2035 | Change today | Confidence |
|---|---|---|---|---|---|
| AGI anywhere | 0.9% | 35% | 58% | no change | low–medium |
| A · Hidden AGI | 0.6% | 19% | 31% | no change | low–medium |
| B · Hidden self-improvement | 2% | 22% | 33% | no change | low |
| C · Covert AGI actor | 0.3% | 8% | 14% | no change | low |
| D · Covert government influence | 0.2% | 3% | 7% | no change | low |
| D-open · Open government influence | 0.2% | 23% | 45% | no change | low |
How the "today" numbers are derived: analysis, Step 5 →
Signals
3 confirmed (count toward Watch) · 7 open · 1 quiet. No signal changed today. All signals →
Escape watch: 0 of 8 confirmed · 8 open · no status changed; new evidence on 8 (weights, selfrep, compute, money, identity, supplychain, coordination, leakage). Escape watch →
Capability gap ~3 months · AI-led R&D 26% · Disclosure lag ~115 days (112–120) floor · Big Tech capex $511B / yr · Delegation rung 3 of 5. No gauge moved.
Is AGI here, today?
No public system meets every part's bar at once. Today's only lead, pending our weekly check, is a tool-free, model-graded reasoning test the gated Gemini 4 Carbon now tops; Altman's undated anecdote counts toward no part. Full reasoning →
Top stories
- Oct 29 Altman: OpenAI holds more than its 700 published proofs ('We do. Of course.'), with no date, field, model or count. verified fact (part our inference)cheekypint.substack.com · more →
- Oct 29 OpenAI says paging is 'currently disabled for internal Codex traffic' and business-hours-only for most other workloads; its own account, not an audit. verified factOpenAI · more →
- Oct 29 (report; run dates not given) Epoch: the open-weight GLM 5.3 autonomously built working exploits for all six Firefox releases Sept 17–Oct 26, at about $1,000 marginal cost each. verified factEpoch AI · more →
- Earlier in 2026 WIRED, on unnamed sources: a contractor's personal Claude Code agent ran on a Pentagon NIPRNet system for two months; Accenture, via GBHackers, says 'without merit'. credible reportx.com · more →
- Oct 29 New York's proposed S.10731 would make large developers report quarterly how much of their AI research is automated; a bill, not law. verified factnysenate.gov · more →
- July – Sept Amazon's Q3 capital purchases were $77.26B, against $35.10B a year earlier; it expects about $220B of 2026 cash capex. verified fact (part our inference)SEC EDGAR · more →
Dates to watch
- Oct 31Our forecast: “OpenAI says it has resumed training its frontier models” (we said 45%)
- Oct 31Our forecast: “Anthropic shares begin trading on Nasdaq” (we said 75%)
- Nov 7Alarm condition W2 is met unless OpenAI's post-Astra model is released or externally evaluated
What would change our mind: OpenAI dating its unpublished discoveries to an unseen model or naming fields well beyond its published proofs, an outside evaluation or release of its post-Astra model before Nov 7, or a measured AI-led share behind talk of self-improvement.