H1 2026 will deliver a 'massive AI breakthrough' driven by unprecedented concentration of computational power at leading American laboratories — AI has been elevated from mere thematic investment to structural 'macro variable' akin to oil prices or int...
Predictor: Morgan Stanley
Prediction text
H1 2026 will deliver a 'massive AI breakthrough' driven by unprecedented concentration of computational power at leading American laboratories — AI has been elevated from mere thematic investment to structural 'macro variable' akin to oil prices or interest rates. | Next GPT-5 / Claude 5 / Gemini 3 capability announcement
Key catalyst: Next GPT-5 / Claude 5 / Gemini 3 capability announcement
Watch events: Frontier-model capability releases; MS AI outlook updates
Resolution evidence
MS 2026 AI outlook publication; GPT-5, Claude 5, Gemini 3 release cycles in H1 2026 align.
Predictor: Morgan Stanley
Calibration plot (stated vs observed)
Evidence about this node from Morgan Stanley 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
This node isn't linked to a reference class. The Bayesian update applies without outside-view blending.
Probability over time
Milestone chain
- 2026-02-22overdueQ1 window check-in (25%)
- 2026-02-28overdueStargate exceeds 10GW milestone with 3GW added in 90 daysHow: OpenAI publicly confirms Stargate datacenter capacity exceeds 10GW with >=3GW added in trailing 90 daysSource: https://openai.com/index/building-the-compute-infrastructure-for-the-intelligence-age/ — Stargate capacity updateconf 92%
- 2026-03-31overdueGPT-5.5 trained at flagship Abilene Stargate siteHow: OpenAI confirms GPT-5.5 trained on NVIDIA GB200 cluster at Stargate Abilene with measurable capability gains over GPT-5.4Source: https://intuitionlabs.ai/articles/openai-stargate-datacenter-details — Stargate Abilene flagshipconf 85%
- 2026-04-16overdueQ2 window check-in (50%)
- 2026-06-08pendingQ3 window check-in (75%)
- 2026-06-30pendingHyperscaler 2026 AI capex hits $700B run-rateHow: Combined Microsoft + Meta + Google + Amazon + Oracle 2026 capex guidance reaches >=$700B annualized rate (vs ~$365B in 2025)Source: https://aijourn.com/700-billion-ai-capex-in-2026-following-the-capital-flows-from-hyperscalers-to-chipmakers/ — $700B capexconf 85%
- 2026-06-01 → 2026-12-31pendingCascade: AI compute treated as macro variable in Fed/Treasury commentaryHow: FOMC minutes, Treasury Quarterly Refunding Statement, or Fed Beige Book reference AI compute concentration as a structural macro driver akin to oil/ratesSource: Pattern extrapolation from St. Louis Fed AI-GDP tracking commentaryconf 55%
- 2026-09-30pendingAI capex contributes 40% of US 2026 GDP growthHow: BEA Q2 2026 advance/second estimate confirms AI-related investment contributed >=40% of US real GDP growth in 1H 2026Source: https://markets.financialcontent.com/stocks/article/marketminute-2026-3-9-ai-supercycle-massive-tech-investments-projected-to-fuel-40-of-us-gdp-growth-in-2026 — AI Supercycle 40%conf 70%
- 2026-07-01 → 2026-12-31pendingOpenAI custom Titan chip mass production at TSMC 3nmHow: OpenAI/Broadcom-designed Titan AI chip enters mass production at TSMC 3nm with first datacenter deployment in H2 2026Source: https://www.datacenterfrontier.com/machine-learning/article/55319132/scaling-stargate-openais-five-new-u-s-data-centers-push-toward-10-gw-ai-infrastructure — Titan chip H2 2026conf 65%
What if this resolves?
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
Raw metadata
{
"trf": 0.5517581791161151,
"kappa": 0.5833,
"base_rate": null,
"predictor": "Morgan Stanley",
"total_llr": -1.6218604324326575,
"grace_days": 7,
"bayesian_v2": true,
"prior_logit": 0,
"bayes_factor": "2.4:1 against",
"blend_reason": "no reference_class linked",
"inside_prior": 0.5,
"kappa_source": "predictor_table",
"n_milestones": 4,
"blend_applied": false,
"contributions": [
{
"llr": -0.4054651081081644,
"kind": "quartile_checkpoint",
"kappa": 0.5833,
"label": "Q1 window check-in (25%)",
"weight": 0.05,
"strength": "weak",
"confidence": null,
"source_url": null,
"adjusted_llr": -0.2365077975594923,
"expected_date": "2026-02-22",
"measurement_criterion": null
},
{
"llr": -0.4054651081081644,
"kind": "llm_pre_event",
"kappa": 0.5366360000000001,
"label": "Stargate exceeds 10GW milestone with 3GW added in 90 days",
"weight": 0.4,
"strength": "weak",
"confidence": 0.92,
"source_url": "https://openai.com/index/building-the-compute-infrastructure-for-the-intelligence-age/",
"adjusted_llr": -0.21758717375473294,
"expected_date": "2026-02-28",
"measurement_criterion": "OpenAI publicly confirms Stargate datacenter capacity exceeds 10GW with >=3GW added in trailing 90 days"
},
{
"llr": -0.4054651081081644,
"kind": "llm_pre_event",
"kappa": 0.495805,
"label": "GPT-5.5 trained at flagship Abilene Stargate site",
"weight": 0.4,
"strength": "weak",
"confidence": 0.85,
"source_url": "https://intuitionlabs.ai/articles/openai-stargate-datacenter-details",
"adjusted_llr": -0.20103162792556845,
"expected_date": "2026-03-31",
"measurement_criterion": "OpenAI confirms GPT-5.5 trained on NVIDIA GB200 cluster at Stargate Abilene with measurable capability gains over GPT-5.4"
},
{
"llr": -0.4054651081081644,
"kind": "quartile_checkpoint",
"kappa": 0.5833,
"label": "Q2 window check-in (50%)",
"weight": 0.05,
"strength": "weak",
"confidence": null,
"source_url": null,
"adjusted_llr": -0.2365077975594923,
"expected_date": "2026-04-16",
"measurement_criterion": null
}
],
"evidence_kind": "metadata_milestone_miss_sweep",
"inside_source": "prior_prob",
"inside_weight": 0.6137692746187193,
"outside_weight": 0.38623072538128067,
"posterior_prob": 0.29077266049571776,
"posterior_logit": -0.8916343967992859,
"predictor_brier": 0.01,
"inside_posterior": 0.29077266049571776,
"blended_posterior": 0.29077266049571776,
"reference_class_id": null,
"total_adjusted_llr": -0.8916343967992859,
"predictor_n_resolved": 1
}Network propagation neighbors
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
Adverse (3)
Prerequisites (2)
| Type | Pred | Title | Domain | Lag |
|---|---|---|---|---|
| correlate | S_NO_AI_PAUSE_5Y | No major AI pause through 2031 | ai_regulatory_pause | — |
| correlate | S_AI_PAUSE_2028 | AI pause beginning 2028 | ai_regulatory_pause | — |
Dependents (0)
| Type | Pred | Title | Domain | Lag |
|---|---|---|---|---|
| No dependents | ||||
Linked documents (10)
Raw metadata
{
"nia": false,
"qty": "H1 2026 breakthrough",
"mode": "FORECAST",
"role": "Cited-Firm",
"context": "Extends CMQ_021 (AI as macro variable) with specific H1 2026 breakthrough timing. Distinct near-term forecast.",
"to_year": 2026,
"conv_cues": "institutional timing commitment",
"direction": "HAPPEN",
"from_year": 2026,
"timeframe": "H1 2026",
"conv_level": "HIGH",
"milestones": [
{
"kind": "quartile_checkpoint",
"label": "Q1 window check-in (25%)",
"status": "overdue",
"weight": 0.05,
"ordinal": -6,
"source_id": null,
"expected_date": "2026-02-22",
"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": "Stargate exceeds 10GW milestone with 3GW added in 90 days",
"source": "https://openai.com/index/building-the-compute-infrastructure-for-the-intelligence-age/ — Stargate capacity update",
"status": "overdue",
"weight": 0.4,
"ordinal": -5,
"source_id": null,
"confidence": 0.92,
"source_url": "https://openai.com/index/building-the-compute-infrastructure-for-the-intelligence-age/",
"expected_date": "2026-02-28",
"miss_emitted_at": "2026-05-02T22:07:21.384228+00:00",
"miss_emitted_by": "metadata_milestone_sweep",
"research_origin": "deep_research",
"measurement_criterion": "OpenAI publicly confirms Stargate datacenter capacity exceeds 10GW with >=3GW added in trailing 90 days"
},
{
"kind": "llm_pre_event",
"label": "GPT-5.5 trained at flagship Abilene Stargate site",
"source": "https://intuitionlabs.ai/articles/openai-stargate-datacenter-details — Stargate Abilene flagship",
"status": "overdue",
"weight": 0.4,
"ordinal": -4,
"source_id": null,
"confidence": 0.85,
"source_url": "https://intuitionlabs.ai/articles/openai-stargate-datacenter-details",
"expected_date": "2026-03-31",
"miss_emitted_at": "2026-05-02T22:07:21.384228+00:00",
"miss_emitted_by": "metadata_milestone_sweep",
"research_origin": "deep_research",
"measurement_criterion": "OpenAI confirms GPT-5.5 trained on NVIDIA GB200 cluster at Stargate Abilene with measurable capability gains over GPT-5.4"
},
{
"kind": "quartile_checkpoint",
"label": "Q2 window check-in (50%)",
"status": "overdue",
"weight": 0.05,
"ordinal": -3,
"source_id": null,
"expected_date": "2026-04-16",
"observed_date": null,
"miss_emitted_at": "2026-05-02T22:07:21.384228+00:00",
"miss_emitted_by": "metadata_milestone_sweep"
},
{
"kind": "quartile_checkpoint",
"label": "Q3 window check-in (75%)",
"status": "pending",
"weight": 0.05,
"ordinal": -2,
"source_id": null,
"expected_date": "2026-06-08",
"observed_date": null
},
{
"kind": "llm_pre_event",
"label": "Hyperscaler 2026 AI capex hits $700B run-rate",
"source": "https://aijourn.com/700-billion-ai-capex-in-2026-following-the-capital-flows-from-hyperscalers-to-chipmakers/ — $700B capex",
"status": "pending",
"weight": 0.4,
"ordinal": -1,
"source_id": null,
"confidence": 0.85,
"source_url": "https://aijourn.com/700-billion-ai-capex-in-2026-following-the-capital-flows-from-hyperscalers-to-chipmakers/",
"expected_date": "2026-06-30",
"research_origin": "deep_research",
"measurement_criterion": "Combined Microsoft + Meta + Google + Amazon + Oracle 2026 capex guidance reaches >=$700B annualized rate (vs ~$365B in 2025)"
},
{
"kind": "event",
"label": "H1 2026 will deliver a 'massive AI breakthrough' driven by unprecedented concentration of computational power at leading American laboratori",
"status": "pending",
"weight": 1,
"ordinal": 0,
"sou
... (truncated)