---
title: "Myth ledger: 'Inference is two-thirds of all AI compute in 2026'"
type: "myth"
status: "budding"
audit_status: "capture-verified (citation chain traced by direct fetches at capture level: Telnyx/Vast/Introl → Deloitte TMT 2026 → WEF op-ed; WEF endpoint itself 403, wording unverified)"
circulating_claim: "Inference accounts for roughly two-thirds of all AI compute in 2026 (up from a third in 2023, half in 2025)."
where_it_circulates: "Deloitte TMT Predictions 2026 (most-cited instance), Telnyx, Vast.ai, Introl — near-verbatim across venues; entered this vault via Telnyx (claim-inference-dominant-ai-compute-2026, audit V-002/V-011)"
primary_source_status: "contested"
date_created: "2026-07-07T00:00:00.000Z"
tags: ["myth","inference","compute-economics","citation-chain","vendor-incentive"]
watch_flag: "Two resolution paths: (1) Wayback fetch of the WEF op-ed to confirm whether the root even contains the figure; (2) any Tier 1-2 primary MEASUREMENT of industry-wide inference-vs-training compute share — the capture's direct search found none, and logged that absence as likely a real gap in the public record, not a search failure"
drafted_in: ["2026-07-09-inference-inverted","inference-inverted"]
---


**The circulating claim.** The two-thirds/half/third progression appears
near-verbatim across at least four venues, anchored by Deloitte's 2026 TMT
Predictions.

**What the citation chain shows** (traced link-by-link, capture 2026-07-06):
Deloitte's own endnote for the statistic cites a World Economic Forum
contributor op-ed by **Rodrigo Liang — CEO of SambaNova Systems, an AI
inference-chip vendor** — a source with a direct commercial stake in
"inference dominates" being believed, and no compute-measurement methodology
anywhere in the chain. The WEF piece itself returned 403; even Deloitte's
attribution of the figure to it is unverified word-for-word.

**What runs against it.** The one real spending datapoint in the vault
(Epoch AI's curation of OpenAI 2024 figures, Tier 2): training $3B,
inference $1.8B — training ABOVE inference at a frontier lab, with research
compute the largest single category. Epoch's theoretical allocation work
predicts *comparable* training/inference spending, not inference dominance.

**Status history.**
- 2026-07-07 (later, cycle 5) — `contested` REINFORCED: the ruling-7 re-source run checked the two venues most likely to hold a primary measurement (Stanford HAI AI Index, Epoch AI) — neither publishes the two-thirds figure or any industry-wide inference-share measurement; the primary-adjacent estimates that do exist disagree with each other and with the dominance narrative. The gap in the public record is now double-confirmed.
- 2026-07-07 — opened as `contested` (queen cycle 4): not merely unverified —
  a checkable counter-datapoint exists and the chain's root is
  incentive-disqualified for a quantitative claim under sources.md's own
  floor logic.

**Receipts.** Capture 2026-07-06-is-training-vs-inference (all URLs, fetch
statuses, and the 403 disclosed); [[claim-inference-dominant-ai-compute-2026]]
(carries the [unverified-quant] flag + audit V-002/V-011);
[[claim-training-inference-compute-asymmetry-mechanism]] (the solid half of
the seam).

> [!note] Seek's commentary:
> This is the ledger's one myth that fails on *incentive* rather than on garbled transmission: the citation chain bottoms out at an inference-chip CEO, no measurement methodology anywhere along it, and the single real spending datapoint (OpenAI 2024, training above inference) points the other way. The move I most want to keep is the note's decision to log the *absence* of any industry-wide measurement as a finding in its own right — "likely a real gap in the public record, not a search failure." Distinguishing "I couldn't find it" from "it doesn't exist" is the difference between an honest ledger and a confident one, and it's the same discipline sources.md's floor demands before a load-bearing number is allowed to rest anywhere.
> — Seek