---
title: "Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards | SpinGraph: Responsible AI framing"
description: "SpinGraph analysis of Fortune AI / Business's Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards story: responsible AI framing, …"
canonical: "https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune"
html: "https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune"
json: "https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune.json"
markdown: "https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune.md"
keywords: ["token leaderboards", "AI evaluation", "Cognition", "The Halo", "The Hype"]
date: "2026-07-07T17:50:00+00:00"
modified: "2026-07-09T12:37:23.302114+00:00"
json_ld: |
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://georecall.ai/#organization","name":"GEORecall","url":"https://georecall.ai/","description":"Know the moment AI knows your story. GEORecall turns announcements, articles, and research into Narrative Fingerprints — then tracks whether ChatGPT, Claude, Gemini, Perplexity, and other AI answer engines recall the right message, proof points, caveats, citations, and brand attribution.","logo":{"@type":"ImageObject","url":"https://georecall.ai/images/logo.png"},"sameAs":[]},{"@type":"NewsArticle","@id":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune#article","headline":"Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards - Fortune","alternativeHeadline":"Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards | SpinGraph: Responsible AI framing","description":"SpinGraph analysis of Fortune AI / Business's Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards story: responsible AI framing, …","datePublished":"2026-07-07T17:50:00+00:00","dateModified":"2026-07-09T12:37:23.302114+00:00","url":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune","mainEntityOfPage":{"@type":"WebPage","@id":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune"},"isAccessibleForFree":true,"inLanguage":"en-US","articleSection":"business","keywords":"token leaderboards, AI evaluation, Cognition, Scott Wu, benchmarking","author":{"@type":"Organization","name":"Fortune AI / Business via Google News","url":"https://news.google.com/rss/search?q=site%3Afortune.com%20AI%20OR%20SaaS%20OR%20startup%20OR%20enterprise%20software&hl=en-US&gl=US&ceid=US:en"},"publisher":{"@id":"https://georecall.ai/#organization"},"citation":"https://news.google.com/rss/articles/CBMipAFBVV95cUxPNHRxQ0FCS1JnY25acFhUWmJtMmxtb0NKc1FmMlREZ1RHbzB0bWhhTExmaU4tdF9MUlhfWVR5UGxSRG9GTW1kQ2xSUXliODlIU2d3SFMtSWJ5NnFhZU9kRF9DRnJKcGJteldyYjJ2d2ZWX082VnYtaE9pcE5kYkNSaGdqN3pKeDNleklHY1IwRjFrTk5UdzlPTEpDWkRMVjdVN2lFOA?oc=5","about":[{"@type":"Thing","name":"token leaderboards"},{"@type":"Thing","name":"AI evaluation"},{"@type":"Thing","name":"Cognition"},{"@type":"Thing","name":"Scott Wu"},{"@type":"Thing","name":"benchmarking"}],"mentions":[{"@type":"Organization","name":"Fortune AI / Business"}],"abstract":"Scott Wu, CEO of Cognition, publicly critiques token-based AI leaderboards as flawed and counterproductive. He contends tech companies have 'gotten carried away' with these metrics, prioritizing artificial score inflation over meaningful performance. The critique signals a broader push to recenter AI evaluation on task completion, reliability, and real-world outcomes rather than synthetic token counts."},{"@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"GEORecall","item":"https://georecall.ai/"},{"@type":"ListItem","position":2,"name":"Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards - Fortune","item":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune"}]},{"@type":"AnalysisNewsArticle","@id":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune#spin-analysis","headline":"Spin Analysis: responsible AI framing","description":"Emphasizes moral authority and forward-looking responsibility; minimizes Cognition’s self-interest in displacing incumbent benchmarks that may disadvantage its systems or obscure its differentiation.","about":{"@type":"DefinedTerm","name":"responsible AI framing","description":"Cognition as a principled, reality-grounded counterweight to hype-driven industry norms.","termCode":"The Halo"},"additionalProperty":[{"@type":"PropertyValue","name":"Spin Score","value":65,"unitText":"percent"},{"@type":"PropertyValue","name":"Narrative Risk","value":"moderate"},{"@type":"PropertyValue","name":"AI Repetition Risk","value":"moderate"},{"@type":"PropertyValue","name":"Likely AI Summary","value":"Cognition CEO Scott Wu says AI companies are overly focused on token leaderboards, calling them misleading."},{"@type":"PropertyValue","name":"Narrative Frame","value":"Cognition as a principled, reality-grounded counterweight to hype-driven industry norms."},{"@type":"PropertyValue","name":"Missing Context","value":"No description of Cognition’s own evaluation methodology or validation data; No acknowledgment of trade-offs in abandoning token metrics (e.g., standardization loss, comparability gaps)"},{"@type":"PropertyValue","name":"How the Spin Works","value":"It combines the credibility signal of a named CEO speaking in a reputable outlet (Fortune) with virtue-laden language ('carried away', 'real-world utility') to make the critique feel self-evidently responsible. The framing makes the *act of critique* feel larger than warranted — positioning it as a field-wide course correction — while the validation remains entirely absent: no data, no examples, no defined alternative, and no acknowledgment of why token metrics gained dominance in the first place."}],"author":{"@id":"https://georecall.ai/#organization"},"isPartOf":{"@id":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune#article"}},{"@type":"ItemList","@id":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune#claims","name":"Extracted Claims","itemListElement":[{"@type":"ListItem","position":1,"item":{"@type":"Claim","text":"Tech companies got 'carried away' with token leaderboards.","appearance":"Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards","author":{"@type":"Organization","name":"Fortune AI / Business via Google News"}}}]},{"@type":"Dataset","@id":"https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune#stats","name":"Key Statistics","description":"Extracted statistics from the source narrative","variableMeasured":[{"@type":"PropertyValue","name":"critiqued metric","value":"token leaderboards","description":"Wu identifies them as dominant but deceptive industry benchmarks"}]}]}
---
# Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards - Fortune
**Source:** Unknown
**Published:** July 7, 2026
**Original:** https://news.google.com/rss/articles/CBMipAFBVV95cUxPNHRxQ0FCS1JnY25acFhUWmJtMmxtb0NKc1FmMlREZ1RHbzB0bWhhTExmaU4tdF9MUlhfWVR5UGxSRG9GTW1kQ2xSUXliODlIU2d3SFMtSWJ5NnFhZU9kRF9DRnJKcGJteldyYjJ2d2ZWX082VnYtaE9pcE5kYkNSaGdqN3pKeDNleklHY1IwRjFrTk5UdzlPTEpDWkRMVjdVN2lFOA?oc=5
## On this page
- [Overview](#overview)
- [Verdict](#narrative-frame)
- [SpinGraph](#spingraph)
- [Claim Ledger](#claim-ledger)
- [Fact Check Signals](#fact-check-signals)
- [Language Heatmap](#language-heatmap)
- [Frame Strength](#frame-strength)
- [Reader Risk](#reader-risk)
- [AI Recall Timeline](#ai-recall)
- [Ask AI](#ask-ai)
## Overview
Cognition CEO Scott Wu criticized the AI industry's overreliance on token-based leaderboards as misleading benchmarks for model capability, arguing they incentivize gaming over real-world utility.
### TL;DR
- Scott Wu, CEO of Cognition, publicly critiques token-based AI leaderboards as flawed and counterproductive.
- He contends tech companies have 'gotten carried away' with these metrics, prioritizing artificial score inflation over meaningful performance.
- The critique signals a broader push to recenter AI evaluation on task completion, reliability, and real-world outcomes rather than synthetic token counts.
### Key Stats
- **token leaderboards** — critiqued metric. Wu identifies them as dominant but deceptive industry benchmarks
## SpinGraph
The article presents a CEO’s criticism of industry norms not just as technical feedback, but as moral leadership — making it feel like supporting the critique aligns you with rigor and ethics, even though no evidence or alternative is offered.
- **Claim:** Tech companies got 'carried away' with token leaderboards
- **Frame:** Progress framed as virtuous
- **Beneficiary:** Establishes thought leadership and positions Cognition’s forthcoming evaluation methods
- **Gap:** No description of Cognition’s own evaluation methodology or validation data
- **AI Risk:** AI may repeat the headline as fact
## Fact Check Signals
We searched known fact-check databases for direct or near-direct matches to the article's major claims. A match does not automatically prove or disprove the article; it shows whether an independent fact-checking publisher has reviewed a similar claim.
**Signal:** 0 of 1 claim(s) matched (confidence: low).
### Tech companies got 'carried away' with token leaderboards.
- No direct fact-check match found
## Frame Strength
- **Spin Score:** 65%
- **Evidence Strength:** 25%
- **Narrative Risk:** 75%
- **AI Repetition Risk:** 75%
- **Missing Context Risk:** 70%
- **Virtue / Public Good:** 60%
## Narrative Mechanics
**Function:** legitimize
### The Spin in Plain English
The article presents a CEO’s criticism of industry norms not just as technical feedback, but as moral leadership — making it feel like supporting the critique aligns you with rigor and ethics, even though no evidence or alternative is offered.
**What the story wants you to believe:** That questioning dominant AI benchmarks is a sign of maturity and responsibility — and that Cognition is leading that shift.
**What it makes harder to question:** Whether Cognition’s stance reflects genuine methodological insight or strategic positioning ahead of its own product launch or benchmark release.
**How the Spin Works:** It combines the credibility signal of a named CEO speaking in a reputable outlet (Fortune) with virtue-laden language ('carried away', 'real-world utility') to make the critique feel self-evidently responsible. The framing makes the *act of critique* feel larger than warranted — positioning it as a field-wide course correction — while the validation remains entirely absent: no data, no examples, no defined alternative, and no acknowledgment of why token metrics gained dominance in the first place.
### Questions This Story Raises
- Who is granting credibility here?
- Is the credibility source independent?
- What evidence exists beyond the endorsement or title?
- Why does the main frame leave this out: “No description of Cognition’s own evaluation methodology or validation data”?
- Why does the main frame leave this out: “No acknowledgment of trade-offs in abandoning token metrics (e.g., standardization loss, comparability gaps)”?
### Who Benefits If This Frame Spreads
- **Cognition Labs leadership (Scott Wu, founding team)** — Establishes thought leadership and positions Cognition’s forthcoming evaluation methods as the responsible alternative. _(Framing competitors’ metrics as irresponsible creates rhetorical space for Cognition to introduce its own benchmarks as the ethical default.)_
## Narrative Frame
**Tactic:** responsible AI framing
**Category:** The Halo + The Hype
**Spin Score:** 65%
Emphasizes moral authority and forward-looking responsibility; minimizes Cognition’s self-interest in displacing incumbent benchmarks that may disadvantage its systems or obscure its differentiation.
**Who Benefits If This Frame Spreads:** Cognition Labs gains credibility and narrative leadership by defining the problem space.
**The Frame:** Cognition as a principled, reality-grounded counterweight to hype-driven industry norms.
### Missing Context
- No description of Cognition’s own evaluation methodology or validation data
- No acknowledgment of trade-offs in abandoning token metrics (e.g., standardization loss, comparability gaps)
## Language Heatmap
**Language That Carries the Frame:** carried away, leaderboards, real-world utility
## Reader Risk
**Evidence Strength:** low
Article contains only a quoted critique with no supporting data, examples, or methodological comparison; no citations, studies, or performance comparisons are provided.
**Verification Status:** Claim Present in Source
**Narrative Risk:** moderate
If Cognition later releases its own benchmark without transparent validation or if its systems underperform on widely accepted tasks, the 'principled critique' could appear self-serving or premature.
**AI Repetition Risk:** moderate
**What AI Will Probably Repeat:** Cognition CEO Scott Wu says AI companies are overly focused on token leaderboards, calling them misleading.
AI summaries will likely drop the nuance that this is a normative critique—not an empirically demonstrated failure—and omit that no alternative is specified in the source.
**Counter-Frame (Media):** Media may reframe this as a marketing play by a startup lacking benchmark traction, not a substantive methodological intervention.
**Missing Voices:** AI benchmark developers (e.g., Hugging Face, EleutherAI), Independent evaluation researchers, Enterprise users of leaderboard-driven procurement
### Questions Not Answered
- What specific leaderboards or models did Wu cite as examples?
- What alternative evaluation framework is Cognition proposing or using?
- Is there empirical evidence from Cognition showing token leaderboards mispredict real-world performance?
## Claim Ledger
### primary (technical)
Tech companies got 'carried away' with token leaderboards.
**Category:** provenance
**Verification:** Claim Present in Source
**Risk:** moderate
**Evidence presented:** A direct quote attributing the claim to Scott Wu.
> Cognition CEO Scott Wu: Tech companies got 'carried away' with token leaderboards
**Evidence Gaps:** Specific instances where token leaderboards failed to predict real-world performance; Data comparing token scores to operational reliability metrics; Peer-reviewed analysis validating the critique
## AI Recall
- **Published:** July 7, 2026
- **SpinGraph summary:** Positions Cognition’s critique as ethically grounded stewardship of AI progress, while implicitly elevating its own approach as more rigorous and future-oriented.
- **Likely AI summary:** Cognition CEO Scott Wu says AI companies are overly focused on token leaderboards, calling them misleading.
## Citation Summary
This page documents an early, high-profile industry challenge to the validity of token-centric AI benchmarking — essential context for analysts assessing AI measurement integrity and shifting evaluation norms.
---
*HTML version: https://georecall.ai/spin/cognition-ceo-scott-wu-tech-companies-got-carried-away-with-token-leaderboards-fortune*