COD_BIO_001predictionBiotech/LongevityFDA-AI-drug-guidance

FDA finalizes or materially advances AI-for-drug-submission guidance by end 2026

Predictor: Codex Research Pack

Prior probability

70.0%

Current probability

46.6%

evolves via intake + LBP

Conviction

4/5

Signal quality

—

Resolution

pending

Window

2025-01-06 – 2026-12-31

Edges in / out

1 / 0

Tickers exposed

Prediction text

FDA finalizes or materially advances AI-for-drug-submission guidance by end 2026

Predictor: Codex Research Pack

κ + Brier as of 2026-06-12

Full calibration →

κ (discount)

0.500

Brier

—

Hits / Misses

0 / 0

Hit rate

—

Evidence about this node from Codex Research Pack 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

4 prob_history rows

intake v2milestone miss sweeplbp propagationreference class assignedlegacy v1prior_prob (analyst seed)current = 46.6%

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: 3 fired ✓ · 4 overdue ⏱

2025-01-07hitFDA draft guidance 'Considerations for Use of AI to Support Regulatory Decision-Making' published Jan 2025
How: FDA Federal Register publishes draft AI-for-drug-submission guidance for industry comment
Source: https://www.federalregister.gov/documents/2025/01/07/2024-31542/considerations-for-the-use-of-artificial-intelligence-to-support-regulatory-decision-making-for-drugconf 99%
Notes: HIT — first FDA guidance on AI for drug submissions; baseline event.
2025-01-07hitFDA proposes 7-step credibility assessment framework for AI models
How: FDA publishes risk-based credibility assessment framework for AI in drug/biological product regulatory decisions
Source: https://www.fda.gov/news-events/press-announcements/fda-proposes-framework-advance-credibility-ai-models-used-drug-and-biological-product-submissionsconf 99%
2025-05-25overdueQ1 window check-in (25%)
2025-10-11overdueQ2 window check-in (50%)
2026-02-15hitIndustry critical reviews of FDA draft published (peer-reviewed)
How: Peer-reviewed academic critical review of FDA AI draft guidance published in indexed journal
Source: https://onlinelibrary.wiley.com/doi/10.1155/joch/5202999conf 95%
Notes: Niazi 2026 critical review in Journal of Chemistry signals industry engagement and pressure for finalization.
2026-02-27overdueQ3 window check-in (75%)
2026-04-01 → 2026-06-30overdueFDA finalizes AI drug-submission guidance by Q2 2026 (per agency timeline)
How: FDA publishes final (non-draft) version of AI drug/biological product submission guidance
Source: https://intuitionlabs.ai/articles/fda-ai-drug-development-guidanceconf 45%
Notes: FDA timeline shows Q2 2026 target; agency timelines often slip 6-12 months.
2026-07-17pendingFDA finalizes or materially advances AI-for-drug-submission guidance by end 2026
2026-06-01 → 2026-12-31pendingMaterially advanced revised draft published if not finalized
How: If not final, FDA publishes revised draft incorporating industry comments OR advances to companion final/draft for biologics specifically
Source: FDA CDER/CBER pipeline trackingconf 70%
Notes: Resolution criterion explicitly allows 'major revised draft/final framework' as HIT.
2026-09-01 → 2027-06-30pendingFirst AI-supported drug NDA approved using credibility framework
How: FDA approves a New Drug Application that explicitly used the AI credibility-assessment framework
Source: FDA approval announcementsconf 40%
Notes: Cascade — would validate that the framework is operational, not just published.

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: 47%)

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

metadata_milestone_miss_sweep2026-05-30T22:15:00Z46.6%-3.9pp

metadata_milestone_miss_sweep bayesian_v2 n=1 inside=0.466 blend=0.466 LLR=-0.155 κ=0.85 no_blend

Raw metadata

{
  "trf": 0.2956808120329126,
  "kappa": 0.8499,
  "base_rate": null,
  "predictor": "Codex Research Pack",
  "total_llr": -0.4054651081081644,
  "grace_days": 7,
  "bayesian_v2": true,
  "prior_logit": 0.019258350505803372,
  "bayes_factor": "1.2:1 against",
  "blend_reason": "no reference_class linked",
  "inside_prior": 0.504814438827484,
  "kappa_source": "predictor_table",
  "n_milestones": 1,
  "blend_applied": false,
  "contributions": [
    {
      "llr": -0.4054651081081644,
      "kind": "llm_pre_event",
      "kappa": 0.382455,
      "label": "FDA finalizes AI drug-submission guidance by Q2 2026 (per agency timeline)",
      "weight": 0.4,
      "strength": "weak",
      "confidence": 0.45,
      "source_url": "https://intuitionlabs.ai/articles/fda-ai-drug-development-guidance",
      "adjusted_llr": -0.155072157921508,
      "expected_date": "2026-05-16",
      "measurement_criterion": "FDA publishes final (non-draft) version of AI drug/biological product submission guidance"
    }
  ],
  "evidence_kind": "metadata_milestone_miss_sweep",
  "inside_source": "history_v2",
  "inside_weight": 0.7930234315769611,
  "outside_weight": 0.2069765684230389,
  "posterior_prob": 0.466098642447458,
  "posterior_logit": -0.13581380741570462,
  "predictor_brier": null,
  "inside_posterior": 0.466098642447458,
  "blended_posterior": 0.466098642447458,
  "reference_class_id": null,
  "total_adjusted_llr": -0.155072157921508,
  "predictor_n_resolved": 0
}

LBP2026-05-10T02:00:02Z50.5%+2.3pp

Network propagation: 48.2% → 50.5%

6-iter LBP, residual 0.00584 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run e5c18d29

LBP2026-05-03T02:00:01Z48.2%+5.9pp

Network propagation: 42.4% → 48.2%

6-iter LBP, residual 0.00677 · damping 0.5, w_intrinsic 0.5 · method lbp_v3 · run 1a683ac9

metadata_milestone_miss_sweep2026-05-02T22:07:21Z42.4%-27.6pp

metadata_milestone_miss_sweep bayesian_v2 n=3 inside=0.424 blend=0.424 LLR=-1.156 κ=0.95 no_blend

Raw metadata

{
  "trf": 0.33436218883920354,
  "kappa": 0.95,
  "base_rate": null,
  "predictor": "Codex Research Pack",
  "total_llr": -1.2163953243244932,
  "grace_days": 7,
  "bayesian_v2": true,
  "prior_logit": 0.8472978603872034,
  "bayes_factor": "3.2:1 against",
  "blend_reason": "no reference_class linked",
  "inside_prior": 0.7,
  "kappa_source": "predictor_table",
  "n_milestones": 3,
  "blend_applied": false,
  "contributions": [
    {
      "llr": -0.4054651081081644,
      "kind": "quartile_checkpoint",
      "kappa": 0.95,
      "label": "Q1 window check-in (25%)",
      "weight": 0.05,
      "strength": "weak",
      "confidence": null,
      "source_url": null,
      "adjusted_llr": -0.38519185270275613,
      "expected_date": "2025-05-25",
      "measurement_criterion": null
    },
    {
      "llr": -0.4054651081081644,
      "kind": "quartile_checkpoint",
      "kappa": 0.95,
      "label": "Q2 window check-in (50%)",
      "weight": 0.05,
      "strength": "weak",
      "confidence": null,
      "source_url": null,
      "adjusted_llr": -0.38519185270275613,
      "expected_date": "2025-10-11",
      "measurement_criterion": null
    },
    {
      "llr": -0.4054651081081644,
      "kind": "quartile_checkpoint",
      "kappa": 0.95,
      "label": "Q3 window check-in (75%)",
      "weight": 0.05,
      "strength": "weak",
      "confidence": null,
      "source_url": null,
      "adjusted_llr": -0.38519185270275613,
      "expected_date": "2026-02-27",
      "measurement_criterion": null
    }
  ],
  "evidence_kind": "metadata_milestone_miss_sweep",
  "inside_source": "prior_prob",
  "inside_weight": 0.7659464678125575,
  "outside_weight": 0.23405353218744251,
  "posterior_prob": 0.42353518888943015,
  "posterior_logit": -0.30827769772106484,
  "predictor_brier": null,
  "inside_posterior": 0.42353518888943015,
  "blended_posterior": 0.42353518888943015,
  "reference_class_id": null,
  "total_adjusted_llr": -1.1555755581082683,
  "predictor_n_resolved": 0
}

Network propagation neighbors

Top edges sorted by latest LBP cross-impact

All propagation →

No propagation data yet. Run inference/.venv/bin/python scripts/ops/run_loopy_belief_propagation.py on the droplet, or wait for the Sunday 02:00 UTC weekly cron.

Ticker exposure

4 ticker(s) linked

Beneficiaries (4)

GOOGL LLY NVS SDGR

Prerequisites (1)

Predictions that must hit first

Type	Pred	Title	Domain	Lag
correlate	247_041	AI-powered drugs have 85% phase 1 success vs 52% traditional	Biotech/Longevity	—

Dependents (0)

Predictions enabled by this

Type	Pred	Title	Domain	Lag
No dependents

Linked documents (10)

Auto-generated by cosine similarity from Polymarket / Manifold / EDGAR / GDELT

Sim	Source	Title	Market prob	Polarity	Reviewed	Published
0.715	fda	FDA ANDA212336: ANDA212336 ((unspecified)) — DR REDDYS LABS LTD	—	mentions	pending	2026-05-07
0.708	fda	FDA ANDA220866: ANDA220866 ((unspecified)) — HETERO LABS LIMITED	—	mentions	pending	2026-06-01
0.700	codex_research_pack	FDA - Framework for AI Models Used for Drug and Biological Product Submissions	—	mentions	pending	2025-01-06
0.698	fda	FDA ANDA214618: LENALIDOMIDE (LENALIDOMIDE) — CIPLA	—	mentions	pending	2026-04-27
0.697	fda	FDA ANDA211182: ANDA211182 ((unspecified)) — HETERO LABS LTD V	—	mentions	pending	2026-04-16
0.695	fda	FDA NDA220442: NDA220442 ((unspecified)) — SHIONOGI INC	—	mentions	pending	2026-05-29
0.692	fda	FDA ANDA213093: ANDA213093 ((unspecified)) — QILU PHARM CO LTD	—	mentions	pending	2026-04-29
0.690	fda	FDA ANDA201452: LENALIDOMIDE (LENALIDOMIDE) — ARROW INTL	—	mentions	pending	2026-04-27
0.689	fda	FDA ANDA216213: LENALIDOMIDE (LENALIDOMIDE) — AMNEAL	—	mentions	pending	2026-04-27
0.688	fda	FDA NDA218197: TRUQAP (CAPIVASERTIB) — ASTRAZENECA	—	mentions	pending	2026-05-27

Raw metadata

From Thesis_Timeline_v1.0_FINAL workbook

{
  "pack_id": "codex_research_event_pack_2026_04_30",
  "milestones": [
    {
      "kind": "llm_pre_event",
      "label": "FDA draft guidance 'Considerations for Use of AI to Support Regulatory Decision-Making' published Jan 2025",
      "notes": "HIT — first FDA guidance on AI for drug submissions; baseline event.",
      "source": "https://www.federalregister.gov/documents/2025/01/07/2024-31542/considerations-for-the-use-of-artificial-intelligence-to-support-regulatory-decision-making-for-drug",
      "status": "hit",
      "weight": 0.4,
      "ordinal": -7,
      "source_id": null,
      "confidence": 0.99,
      "source_url": "https://www.federalregister.gov/documents/2025/01/07/2024-31542/considerations-for-the-use-of-artificial-intelligence-to-support-regulatory-decision-making-for-drug",
      "expected_date": "2025-01-07",
      "observed_date": "2025-01-07",
      "hit_emitted_at": "2026-06-08T22:11:23.030711+00:00",
      "research_origin": "deep_research",
      "measurement_criterion": "FDA Federal Register publishes draft AI-for-drug-submission guidance for industry comment"
    },
    {
      "kind": "llm_pre_event",
      "label": "FDA proposes 7-step credibility assessment framework for AI models",
      "source": "https://www.fda.gov/news-events/press-announcements/fda-proposes-framework-advance-credibility-ai-models-used-drug-and-biological-product-submissions",
      "status": "hit",
      "weight": 0.4,
      "ordinal": -6,
      "source_id": null,
      "confidence": 0.99,
      "source_url": "https://www.fda.gov/news-events/press-announcements/fda-proposes-framework-advance-credibility-ai-models-used-drug-and-biological-product-submissions",
      "expected_date": "2025-01-07",
      "observed_date": "2025-01-07",
      "hit_emitted_at": "2026-06-08T22:11:23.030711+00:00",
      "research_origin": "deep_research",
      "measurement_criterion": "FDA publishes risk-based credibility assessment framework for AI in drug/biological product regulatory decisions"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q1 window check-in (25%)",
      "status": "overdue",
      "weight": 0.05,
      "ordinal": -5,
      "source_id": null,
      "expected_date": "2025-05-25",
      "observed_date": null,
      "miss_emitted_at": "2026-05-02T22:07:21.384228+00:00",
      "miss_emitted_by": "metadata_milestone_sweep"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q2 window check-in (50%)",
      "status": "overdue",
      "weight": 0.05,
      "ordinal": -4,
      "source_id": null,
      "expected_date": "2025-10-11",
      "observed_date": null,
      "miss_emitted_at": "2026-05-02T22:07:21.384228+00:00",
      "miss_emitted_by": "metadata_milestone_sweep"
    },
    {
      "kind": "llm_pre_event",
      "label": "Industry critical reviews of FDA draft published (peer-reviewed)",
      "notes": "Niazi 2026 critical review in Journal of Chemistry signals industry engagement and pressure for finalization.",
      "source": "https://onlinelibrary.wiley.com/doi/10.1155/joch/5202999",
      "status": "hit",
      "weight": 0.4,
      "ordinal": -3,
      "source_id": null,
      "confidence": 0.95,
      "source_url": "https://onlinelibrary.wiley.com/doi/10.1155/joch/5202999",
      "expected_date": "2026-02-15",
      "observed_date": "2026-02-15",
      "hit_emitted_at": "2026-06-08T22:11:23.030711+00:00",
      "research_origin": "deep_research",
      "measurement_criterion": "Peer-reviewed academic critical review of FDA AI draft guidance published in indexed journal"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q3 window check-in (75%)",
      "status": "overdue",
      "weight": 0.05,
      "ordinal": -2,
      "source_id": null,
      "expected_date": "2026-02-27",
      "observed_date": null,
      "miss_emitted_at": "2026-05-02T22:07:21.384228+00:00",
      "miss_emitted_by": "metadata_milestone_sweep"
    },
    {
      "kind": "llm_pre_event",
      "l
... (truncated)