← Cockpit
230_046predictionAIAI-scaling

OpenAI's Noam Brown has only ~3 months of future model access internally — dramatic capability gap with public.

Predictor: Dave Blundin · ep#230 "AI CEOs Come Online: Sam Altman's Replacement Plan, Job Loss & 'Solve Everything' Launches |EP #230" · source

Prior probability
60.0%
Current probability
43.3%
evolves via intake + LBP
Conviction
4/5
Signal quality
B
Resolution
pending
Window
2026-01-01 – 2026-08-31
Edges in / out
10 / 5
Tickers exposed
37

Prediction text

OpenAI's Noam Brown has only ~3 months of future model access internally — dramatic capability gap with public. | Well, and even today, you know, if you talk to Noam Brown over at OpenAI, you know, he's working on the next generation internally, but it's only like three months in the future that, you know, he has access to. But 3 months in the future in the era of self-improvement is like you know massively different intelligence level.

Watch events: OpenAI next funding round; IPO timing; revenue disclosures

Verbatim quote

From episode "AI CEOs Come Online: Sam Altman's Replacement Plan, Job Loss & 'Solve Everything' Launches |EP #230"
Well, and even today, you know, if you talk to Noam Brown over at OpenAI, you know, he's working on the next generation internally, but it's only like three months in the future that, you know, he has access to. But 3 months in the future in the era of self-improvement is like you know massively different intelligence level.

Predictor: Dave Blundin

κ + Brier as of 2026-05-22
κ (discount)
0.821
Brier
0.0491
excellent
Hits / Misses
3 / 2
of 9 resolved
Hit rate
33.3%
Calibration plot (stated vs observed)

Evidence about this node from Dave Blundin is multiplied by κ in /api/intake. Lower κ = less weight; floors at 0.10 (effectively silenced) and caps at 1.00 (full weight).

Reference class

Not linked

This node isn't linked to a reference class. The Bayesian update applies without outside-view blending.

Probability over time

7 prob_history rows
0%25%50%75%100%prior 60%2026-04-302026-05-102026-05-24
intake v2milestone miss sweeplbp propagationreference class assignedlegacy v1prior_prob (analyst seed)current = 43.3%

Milestone chain

Pre-event signals (upstream prereqs + window checkpoints) → resolution event → downstream cascades. Status/dates update from linked nodes; re-derive nightly via scripts/ops/derive_milestones.py.
Leading chain: 4 fired ✓ · 3 overdue ⏱ · 1 pending
  1. 2026-01-01 → 2026-08-31overdueFrontier model release cadence remains <=4 months between major versions
    How: OpenAI ships 2+ frontier model upgrades within calendar 2026 (Jan-Aug window), each with new capability tier disclosure, validating Brown's '~3 months ahead' framing
    Source: https://releasebot.io/updates/openai — OpenAI release notes May 2026conf 85%
  2. 2026-01-01 → 2026-08-31overdueOpenAI/Anthropic safety evaluation period >=4 weeks
    How: Public model card or system card discloses pre-release safety evaluation window of 4-12 weeks between training completion and public release, supporting Brown's gap claim
    Source: OpenAI system cards, Anthropic safety reportsconf 75%
    Notes: Brown's '3 months' gap is consistent with publicly disclosed eval-then-deploy cycles.
  3. 2026-01-01 → 2026-08-31overdueInternal-only model demos at OpenAI DevDay or research blog
    How: OpenAI publicly demonstrates an unreleased model with capabilities clearly above shipped frontier (e.g., new benchmark tier) at DevDay, research preview, or red-team announcement
    Source: OpenAI DevDay 2026, openai.com/researchconf 70%
  4. 2026-01-01 → 2026-08-31pendingFrontier benchmark advances (FrontierMath, ARC-AGI) signal capability jumps every quarter
    How: FrontierMath, ARC-AGI-2, GPQA-Diamond, or SWE-bench-Verified shows >=2 step changes (>10pp absolute gain) within Jan-Aug 2026 window, consistent with rapid internal-public roll-forward
    Source: FrontierMath leaderboard, ARC Prize 2026conf 80%
    Notes: If self-improvement narrative holds, capability gaps should compound quarterly.
  5. 2026-06-01 → 2026-08-31pendingOpenAI ships next-generation reasoning model in mid-2026
    How: OpenAI releases successor to GPT-5/o-series model (e.g., GPT-5.5, o3-pro, or new reasoning architecture) with publicly disclosed capability uplift
    Source: https://www.trendingtopics.eu/openais-gpt-5-5-is-about-to-launch-soon/ — GPT-5.5 imminent launchconf 85%

What if this resolves?

Clamp this prediction TRUE or FALSE and run a counterfactual Gibbs sample. Surfaces the predictions whose marginals shift most under that assumption.
(live posterior: 43%)

Click a button to clamp this prediction and run a Gibbs sample. Returns the predictions whose marginals shift most. ~30s per run; ideal for stress-testing "if X resolves, what else moves?"

Evidence chain

Every probability update with full Bayesian provenance — chronological, latest first
LBP2026-05-24T02:00:02Z43.3%+4.1pp
Network propagation: 39.2% → 43.3%
4-iter LBP, residual 0.01000 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run 806b02f8
LBP2026-05-17T02:00:01Z39.2%+7.8pp
Network propagation: 31.3% → 39.2%
5-iter LBP, residual 0.00689 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run e607fa96
metadata_milestone_miss_sweep2026-05-10T22:10:52Z31.3%-18.2pp
metadata_milestone_miss_sweep bayesian_v2 n=3 inside=0.313 blend=0.313 LLR=-0.766 κ=0.82 no_blend
Raw metadata
{
  "trf": 0.46312307053269436,
  "kappa": 0.8214,
  "base_rate": null,
  "predictor": "Dave Blundin",
  "total_llr": -1.2163953243244932,
  "grace_days": 7,
  "bayesian_v2": true,
  "prior_logit": -0.01913257734403183,
  "bayes_factor": "2.2:1 against",
  "blend_reason": "no reference_class linked",
  "inside_prior": 0.4952170015666818,
  "kappa_source": "predictor_table",
  "n_milestones": 3,
  "blend_applied": false,
  "contributions": [
    {
      "llr": -0.4054651081081644,
      "kind": "llm_pre_event",
      "kappa": 0.69819,
      "label": "Frontier model release cadence remains <=4 months between major versions",
      "weight": 0.4,
      "strength": "weak",
      "confidence": 0.85,
      "source_url": "https://releasebot.io/updates/openai",
      "adjusted_llr": -0.2830916838300393,
      "expected_date": "2026-05-02",
      "measurement_criterion": "OpenAI ships 2+ frontier model upgrades within calendar 2026 (Jan-Aug window), each with new capability tier disclosure, validating Brown's '~3 months ahead' framing"
    },
    {
      "llr": -0.4054651081081644,
      "kind": "llm_pre_event",
      "kappa": 0.61605,
      "label": "OpenAI/Anthropic safety evaluation period >=4 weeks",
      "weight": 0.4,
      "strength": "weak",
      "confidence": 0.75,
      "source_url": null,
      "adjusted_llr": -0.24978677985003467,
      "expected_date": "2026-05-02",
      "measurement_criterion": "Public model card or system card discloses pre-release safety evaluation window of 4-12 weeks between training completion and public release, supporting Brown's gap claim"
    },
    {
      "llr": -0.4054651081081644,
      "kind": "llm_pre_event",
      "kappa": 0.5749799999999999,
      "label": "Internal-only model demos at OpenAI DevDay or research blog",
      "weight": 0.4,
      "strength": "weak",
      "confidence": 0.7,
      "source_url": null,
      "adjusted_llr": -0.23313432786003233,
      "expected_date": "2026-05-02",
      "measurement_criterion": "OpenAI publicly demonstrates an unreleased model with capabilities clearly above shipped frontier (e.g., new benchmark tier) at DevDay, research preview, or red-team announcement"
    }
  ],
  "evidence_kind": "metadata_milestone_miss_sweep",
  "inside_source": "history_v2",
  "inside_weight": 0.675813850627114,
  "outside_weight": 0.32418614937288603,
  "posterior_prob": 0.3132120021709241,
  "posterior_logit": -0.7851453688841381,
  "predictor_brier": 0.0491,
  "inside_posterior": 0.3132120021709241,
  "blended_posterior": 0.3132120021709241,
  "reference_class_id": null,
  "total_adjusted_llr": -0.7660127915401063,
  "predictor_n_resolved": 9
}
LBP2026-05-10T02:00:02Z49.5%-1.2pp
Network propagation: 50.7% → 49.5%
6-iter LBP, residual 0.00584 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run e5c18d29
LBP2026-05-03T02:00:01Z50.7%-2.2pp
Network propagation: 52.9% → 50.7%
6-iter LBP, residual 0.00677 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run 1a683ac9
LBP2026-04-30T16:39:51Z52.9%-2.9pp
Network propagation: 55.8% → 52.9%
5-iter LBP, residual 0.00825 · damping 0.5, w_intrinsic 0.5 · method lbp_v2 · run 0c8a4ea3
LBP2026-04-30T02:18:57Z55.8%-4.2pp
Network propagation: 60.0% → 55.8%
5-iter LBP, residual 0.00825 · damping 0.5, w_intrinsic 0.5 · method lbp_v1 · run 592311ef

Network propagation neighbors

Top edges sorted by latest LBP cross-impact
All propagation →

Top incoming (parents)

Edges that influence THIS node's belief

KindNodeTheir probP(c|s=T)P(c|s=F)Δ implied
killerTK03
AI Regulatory Moratorium (EU/US Capability Freeze)
10.0%0.0500.600+0.112
killerTK02
AI Compute Supply Shock (TSMC/Taiwan Disruption)
12.0%0.0500.600+0.101
prereqSEM_014
Nvidia's Arizona-based TSMC factory successfully fabricated Jensen Huang
86.1%0.6000.050+0.086
killerTK01
AGI Capability Plateau (2026-27 Training Stall)
15.0%0.0500.600+0.085
prereqSEM_011
Nvidia became the world's first $5 trillion company (late 20Jensen Huang
85.5%0.6000.050+0.084

Top outgoing (children)

Predictions THIS node influences

KindNodeTheir probP(c|s=T)P(c|s=F)Δ implied
prereq247_023
AI will be able to do everything a white collar worker does Dave Blundin
40.8%0.7200.050-0.052
prereq242_031
Most large companies' business models will be disrupted in 2Peter Diamandis
36.1%0.6500.050-0.038
prereq244_019
Peter's son won't need a driver's license in 2 yearsPeter Diamandis
48.4%0.9200.050-0.038
prereq230_020
Peter's 14-year-old son Milan will never get a driver's licePeter Diamandis
34.7%0.6500.050-0.023
prereq232_055
We're exiting the industrial age permanently as recursive sePeter Diamandis
35.5%0.7000.050-0.008

Ticker exposure

37 ticker(s) linked

Beneficiaries (24)

MUWULFIRENEQIXALABAPLDASMIYASMLPLABNVDANBISCRWVAAPLAMTAMZNDELLGOOGLIRMLNVGYMETAMSFTORCLSFTBYSTX

Adverse (6)

ACNGENCHGGIBMWNSLRN

Prerequisites (10)

Predictions that must hit first
TypePredTitleDomainLag
prereqSEM_011Nvidia became the world's first $5 trillion company (late 2025), operating a near-monopoly on advanced AI chips.Capital Markets
prereqSEM_027Nvidia Data Center revenue +66% YoY, contributing ~90% of $57B fiscal Q3 revenue; >$4.5T market cap entirely underpinned by AI silicon.Capital Markets
prereqSEM_014Nvidia's Arizona-based TSMC factory successfully fabricated cutting-edge semiconductors on US soil for first time in decades (October 2025).Manufacturing
prereqSEM_012Nvidia quadrupled chip production output while only doubling human headcount — achieved by deploying AI coding tools (Cursor, Claude Code) across engineering.AI/Manufacturing
prereqSEM_015Nvidia agreed to remit 15% of China chip-sale revenue directly to US government in exchange for reversing specific AI chip export bans.Policy/Semis
killerTK09Energy Grid Cap (Data Center Power Wall)
killerTK05Rate Regime Persistence (10y > 5% through 2028)
killerTK01AGI Capability Plateau (2026-27 Training Stall)
killerTK02AI Compute Supply Shock (TSMC/Taiwan Disruption)
killerTK03AI Regulatory Moratorium (EU/US Capability Freeze)

Dependents (5)

Predictions enabled by this
TypePredTitleDomainLag
prereq244_019Peter's son won't need a driver's license in 2 yearsAuto/Transport
prereq247_023AI will be able to do everything a white collar worker does imminentlyAI
prereq232_055We're exiting the industrial age permanently as recursive self-improvement unfolds.AI
prereq242_031Most large companies' business models will be disrupted in 2-5 yearsMarkets/Stocks
prereq230_020Peter's 14-year-old son Milan will never get a driver's license.Auto/Transport

Linked documents (6)

Auto-generated by cosine similarity from Polymarket / Manifold / EDGAR / GDELT
SimSourceTitleMarket probPolarityReviewedPublished
0.640manifoldWill Apple's WWDC 2026 keynote include any verbal reference to OpenAI?39%mentionspending2026-05-16
0.631github_releaseopenai/openai-python v2.33.0mentionspending2026-04-28
0.617github_releaseopenai/openai-python v2.40.0mentionspending2026-06-01
0.616github_releaseopenai/openai-python v2.35.1mentionspending2026-05-06
0.612github_releaseopenai/openai-python v2.34.0mentionspending2026-05-04
0.605github_releaseopenai/openai-python v2.7.1mentionspending2025-11-04

Raw metadata

From Thesis_Timeline_v1.0_FINAL workbook
{
  "nia": false,
  "qty": "3 months",
  "url": "https://www.youtube.com/watch?v=6P0uTDGDr-I",
  "mode": "THESIS",
  "role": "Host",
  "context": "it's only like three months in the future that, you know, he has access to. But 3 months in the future in the era of self-improvement is like you know massively different intelligence level.",
  "to_year": 2026,
  "verbatim": "Well, and even today, you know, if you talk to Noam Brown over at OpenAI, you know, he's working on the next generation internally, but it's only like three months in the future that, you know, he has access to. But 3 months in the future in the era of self-improvement is like you know massively different intelligence level.",
  "conv_cues": "it's now or never",
  "direction": "HAPPEN",
  "from_year": 2026,
  "timeframe": "now (3 months ahead)",
  "conv_level": "HIGH",
  "milestones": [
    {
      "kind": "prereq",
      "label": "Nvidia became the world's first $5 trillion company (late 2025), operating a near-monopoly on advanced AI chips.",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -8,
      "source_id": "SEM_011",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "prereq",
      "label": "Nvidia Data Center revenue +66% YoY, contributing ~90% of $57B fiscal Q3 revenue; >$4.5T market cap entirely underpinned by AI silicon.",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -7,
      "source_id": "SEM_027",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "prereq",
      "label": "Nvidia's Arizona-based TSMC factory successfully fabricated cutting-edge semiconductors on US soil for first time in decades (October 2025).",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -6,
      "source_id": "SEM_014",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "prereq",
      "label": "Nvidia quadrupled chip production output while only doubling human headcount — achieved by deploying AI coding tools (Cursor, Claude Code) a",
      "status": "hit",
      "weight": 0.5,
      "ordinal": -5,
      "source_id": "SEM_012",
      "expected_date": "2026-04-29",
      "observed_date": "2026-04-29"
    },
    {
      "kind": "llm_pre_event",
      "label": "Frontier model release cadence remains <=4 months between major versions",
      "source": "https://releasebot.io/updates/openai — OpenAI release notes May 2026",
      "status": "overdue",
      "weight": 0.4,
      "ordinal": -4,
      "source_id": null,
      "confidence": 0.85,
      "source_url": "https://releasebot.io/updates/openai",
      "expected_date": "2026-05-02",
      "miss_emitted_at": "2026-05-10T22:10:52.342846+00:00",
      "miss_emitted_by": "metadata_milestone_sweep",
      "research_origin": "deep_research",
      "expected_date_range": {
        "to": "2026-08-31",
        "from": "2026-01-01"
      },
      "measurement_criterion": "OpenAI ships 2+ frontier model upgrades within calendar 2026 (Jan-Aug window), each with new capability tier disclosure, validating Brown's '~3 months ahead' framing"
    },
    {
      "kind": "llm_pre_event",
      "label": "OpenAI/Anthropic safety evaluation period >=4 weeks",
      "notes": "Brown's '3 months' gap is consistent with publicly disclosed eval-then-deploy cycles.",
      "source": "OpenAI system cards, Anthropic safety reports",
      "status": "overdue",
      "weight": 0.4,
      "ordinal": -3,
      "source_id": null,
      "confidence": 0.75,
      "expected_date": "2026-05-02",
      "miss_emitted_at": "2026-05-10T22:10:52.342846+00:00",
      "miss_emitted_by": "metadata_milestone_sweep",
      "research_origin": "training",
      "expected_date_range": {
        "to": "2026-08-31",
        "from": "2026-01-01"
      },
      "measurement_criterion": "Public model card or system card discloses pre-release safety evaluation window of 4-12 weeks between trai
... (truncated)