← Cockpit
CMQ_044predictionAI/ComputeCPU-GPU-ratio

Future data-center architectures optimized for agentic workflows may require 1:2 or even 2:1 CPU-to-GPU ratio (vs historical 1:12) to prevent GPU idle-waiting.

Predictor: Morgan Stanley

Prior probability
65.0%
Current probability
65.0%
evolves via intake + LBP
Conviction
5/5
Signal quality
A
Resolution
pending
Window
2026-01-01 – 2030-12-31
Edges in / out
1 / 0
Tickers exposed
31

Prediction text

Future data-center architectures optimized for agentic workflows may require 1:2 or even 2:1 CPU-to-GPU ratio (vs historical 1:12) to prevent GPU idle-waiting. | Agentic rack architecture announcements

Key catalyst: Agentic rack architecture announcements

Watch events: Agentic server SKU architectures; Grace Blackwell adoption; AMD EPYC AI-rack design wins.

Resolution evidence

Status: pending

NVIDIA Grace Blackwell integrates 2:1 CPU:GPU approximately; hyperscaler agentic-rack designs emerging 2026.

Predictor: Morgan Stanley

κ + Brier as of 2026-05-22
κ (discount)
0.633
Brier
0.0442
excellent
Hits / Misses
1 / 0
of 2 resolved
Hit rate
50.0%
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

Not linked

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

Probability over time

0 prob_history rows
No probability history yet.

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: 7 pending
  1. 2026-12-07pendingQ1 window check-in (25%)
  2. 2026-06-01 → 2027-12-31pendingMajor hyperscaler announces agentic-rack architecture with disclosed CPU:GPU ratio >=1:2
    How: AWS re:Invent, Google Cloud Next, MSFT Build, or NVIDIA GTC keynote unveils named server/rack SKU specifically targeting agentic workflows with CPU:GPU ratio of 1:2 or denser CPU side
    Source: https://www.trendforce.com/research/download/RP260408ADconf 65%
  3. 2026-09-01 → 2028-06-30pendingIntel/AMD/ARM CPU vendor reports orchestration-CPU revenue tied to agentic workloads as named segment
    How: Intel Xeon, AMD EPYC, or NVIDIA Grace earnings call breaks out 'agentic orchestration' or 'AI control plane' as a disclosed revenue category >=$1B run-rate
    Source: https://aninews.in/news/business/morgan-stanley-agentic-ai-shifts-value-from-gpus-to-cpus-and-memory-creating-up-to-60bn-incremental-cpu-tam-by-203020260422131744/conf 60%
  4. 2027-11-13pendingQ2 window check-in (50%)
  5. 2027-06-01 → 2029-12-31pendingIndustry-wide cluster-level CPU:GPU ratio shift confirmed at 1:1-1:2 in TrendForce/IDC data
    How: TrendForce, IDC, or Omdia annual data-center hardware report records average new-build CPU:GPU ratio for AI clusters at 1:1 or 1:2 (vs historical 1:4-1:12)
    Source: https://insights.trendforce.com/p/agentic-ai-cpu-gpuconf 55%
  6. 2028-10-18pendingQ3 window check-in (75%)
  7. 2028-01-01 → 2030-12-31pendingOrchestration-CPU TAM tracking toward Morgan Stanley $82.5B-$110B 2030 projection
    How: Morgan Stanley or comparable sell-side update confirms data-center orchestration CPU TAM is on trajectory to exceed $80B by 2030
    Source: https://finance.biggo.com/news/8msmrp0BrdTHlKtC3vWNconf 50%

No downstream cascades — this prediction is a leaf in the dependency graph.

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: 65%)

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

No probability history yet. The first evidence will arrive via /api/intake or the daily milestone sweep / weekly LBP run.

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

31 ticker(s) linked

Beneficiaries (24)

ALABAMBAAPLDARMCEVACRWVDOCNIRENNBISNVDASITMTSMCSCODELLQCOMLNVGYAMDANETMRVLSIEGYAVGONXPIGOOGLINTC

Prerequisites (1)

Predictions that must hit first
TypePredTitleDomainLag
correlateS_COMPUTE_100GW_2030Compute: 100GW national-scale by Dec 2030compute_scale

Dependents (0)

Predictions enabled by this
TypePredTitleDomainLag
No dependents

Linked documents (10)

Auto-generated by cosine similarity from Polymarket / Manifold / EDGAR / GDELT
SimSourceTitleMarket probPolarityReviewedPublished
0.700arxivDon't Let a Few Network Failures Slow the Entire AllReducementionspending2026-06-01
0.688arxivEfficient Serving for Dynamic Agent Workflows with Prediction-based KV-Cache Managementmentionspending2026-05-07
0.684arxivCoral: Cost-Efficient Multi-LLM Serving over Heterogeneous Cloud GPUsmentionspending2026-05-05
0.676arxivA Workflow-Oriented Framework for Asynchronous Human-AI Collaboration in Hybrid and Compute-Intensive HPC Environmentsmentionspending2026-05-05
0.667arxivA Readiness-Driven Runtime for Pipeline-Parallel Training under Runtime Variabilitymentionspending2026-05-18
0.657arxivCaliper-in-the-Loop: Black-Box Optimization for Hyperledger Fabric Performance Tuningmentionspending2026-05-04
0.650arxivGuard: Scalable Straggler Detection and Node Health Management for Large-Scale Trainingmentionspending2026-05-18
0.634github_releasefacebookresearch/faiss v1.6.4mentionspending2020-10-22
0.634github_releasetensorflow/tensorflow v2.17.0-rc1mentionspending2024-07-02
0.633github_releasetensorflow/tensorflow v2.17.0-rc0mentionspending2024-06-18

Raw metadata

From Thesis_Timeline_v1.0_FINAL workbook
{
  "nia": false,
  "qty": "1:2 to 2:1 CPU:GPU",
  "mode": "FORECAST",
  "role": "Cited-Firm",
  "context": "Architectural inversion: CPU-heavy agentic racks; massive re-sizing of server CPU TAM.",
  "to_year": 2030,
  "cited_by": "Synthesis report",
  "conv_cues": "specific ratio; major bank",
  "direction": "NUMERIC_TARGET",
  "from_year": 2026,
  "timeframe": "2026-2030",
  "conv_level": "HIGH",
  "milestones": [
    {
      "kind": "quartile_checkpoint",
      "label": "Q1 window check-in (25%)",
      "status": "pending",
      "weight": 0.05,
      "ordinal": -7,
      "source_id": null,
      "expected_date": "2026-12-07",
      "observed_date": null
    },
    {
      "kind": "llm_pre_event",
      "label": "Major hyperscaler announces agentic-rack architecture with disclosed CPU:GPU ratio >=1:2",
      "source": "https://www.trendforce.com/research/download/RP260408AD",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -6,
      "source_id": null,
      "confidence": 0.65,
      "expected_date": "2027-03-17",
      "research_origin": "training",
      "expected_date_range": {
        "to": "2027-12-31",
        "from": "2026-06-01"
      },
      "measurement_criterion": "AWS re:Invent, Google Cloud Next, MSFT Build, or NVIDIA GTC keynote unveils named server/rack SKU specifically targeting agentic workflows with CPU:GPU ratio of 1:2 or denser CPU side"
    },
    {
      "kind": "llm_pre_event",
      "label": "Intel/AMD/ARM CPU vendor reports orchestration-CPU revenue tied to agentic workloads as named segment",
      "source": "https://aninews.in/news/business/morgan-stanley-agentic-ai-shifts-value-from-gpus-to-cpus-and-memory-creating-up-to-60bn-incremental-cpu-tam-by-203020260422131744/",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -5,
      "source_id": null,
      "confidence": 0.6,
      "expected_date": "2027-08-01",
      "research_origin": "training",
      "expected_date_range": {
        "to": "2028-06-30",
        "from": "2026-09-01"
      },
      "measurement_criterion": "Intel Xeon, AMD EPYC, or NVIDIA Grace earnings call breaks out 'agentic orchestration' or 'AI control plane' as a disclosed revenue category >=$1B run-rate"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q2 window check-in (50%)",
      "status": "pending",
      "weight": 0.05,
      "ordinal": -4,
      "source_id": null,
      "expected_date": "2027-11-13",
      "observed_date": null
    },
    {
      "kind": "llm_pre_event",
      "label": "Industry-wide cluster-level CPU:GPU ratio shift confirmed at 1:1-1:2 in TrendForce/IDC data",
      "source": "https://insights.trendforce.com/p/agentic-ai-cpu-gpu",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -3,
      "source_id": null,
      "confidence": 0.55,
      "expected_date": "2028-09-15",
      "research_origin": "training",
      "expected_date_range": {
        "to": "2029-12-31",
        "from": "2027-06-01"
      },
      "measurement_criterion": "TrendForce, IDC, or Omdia annual data-center hardware report records average new-build CPU:GPU ratio for AI clusters at 1:1 or 1:2 (vs historical 1:4-1:12)"
    },
    {
      "kind": "quartile_checkpoint",
      "label": "Q3 window check-in (75%)",
      "status": "pending",
      "weight": 0.05,
      "ordinal": -2,
      "source_id": null,
      "expected_date": "2028-10-18",
      "observed_date": null
    },
    {
      "kind": "llm_post_event",
      "label": "Orchestration-CPU TAM tracking toward Morgan Stanley $82.5B-$110B 2030 projection",
      "source": "https://finance.biggo.com/news/8msmrp0BrdTHlKtC3vWN",
      "status": "pending",
      "weight": 0.4,
      "ordinal": -1,
      "source_id": null,
      "confidence": 0.5,
      "expected_date": "2029-07-01",
      "research_origin": "training",
      "expected_date_range": {
        "to": "2030-12-31",
        "from": "2028-01-01"
      },
      "measurement_criterion": "Morgan Stanley or compar
... (truncated)