AI Hype Tracker
A site that keeps score. Longitudinal claim tracking, a transparent weekly index, and a daily digest from a deliberately balanced source diet.
HYPE INDEX WEEK 2026-20
methodology →–
A single number 0–100 capturing how hyped the AI conversation is this week. Built from four interrogable components below, each scored 0–25. Higher means more capability-claim volume, funding pressure, and mainstream media saturation — and less skeptic friction (the index is inverted on that component, so a noisy skeptic press lowers the score).
Trend chart populates as weeks accumulate.
GPT‑5 launch commentary spiked ‘phase change’ language; eval dispersion tempered the tails.
Mega-cap capex guides held elevated; vendor financing and GPU backlog stories stayed on page one.
Mainstream front pages ran AI explainers tied to GDP, jobs, and ‘sci-fi’ ledes—attention, not evidence.
Bearish equity notes and ROI skepticism were audible but not dominant; friction rose modestly vs the prior week.
DIGEST · SUNDAY, MAY 17
read full →- The New York Times — The New York Times
- The Economist — The Economist
- Protocol — Protocol
RECENT ESSAYS · 38 IN TOTAL
all essays →Policy & Regulation
The high-risk deadline that moved: what the EU AI Act actually required on 2 August 2026
The date Brussels spent two years pointing at came and went with the high-risk regime deferred to 2027 and 2028. What did land, what slipped, and what the deferral says about regulating on a fixed calendar.
Industry & Investment
Q2 2026: the quarter AI capex reached the cash flow statement
Alphabet posted negative free cash flow for the first time as a public company, Amazon's trailing-twelve-month free cash flow went negative, and two of the four biggest reported profits were driven by unrealized gains. What the filings actually say.
Hype vs Reality
Doubling every four months: auditing the time-horizon claim
METR's task-completion time horizon became 2026's favourite capability statistic. What the underlying measurement supports, what its own authors say it doesn't, and why the newest models are breaking the instrument.
Hype vs Reality
GPT-5 release: capability deltas vs the narrative
Measured comparison of what shipped against the pre-release framing — and why the "phase transition" rhetoric mostly didn't survive contact with the benchmarks.
Model Comparisons
Reasoning models — o1 → o3 → DeepSeek R1 → Claude Opus 4.x thinking
What's actually new in the reasoning-model wave, where the capability ceilings sit, and which benchmarks are starting to get gamed.
Hype vs Reality
Agentic coding: Cursor, Devin, Claude Code, Replit Agent — adoption data vs marketing decks
Where the published adoption metrics actually land for each agentic coding product, and what gets quietly conflated when vendors talk "AI software engineer."
COMPANIES · 10
Anthropic · Cohere · Cursor · DeepSeek · Google DeepMind · Meta AI (FAIR) · Mistral AI · OpenAI · Perplexity · xAI
MODELS · 19
Claude 3.5 Sonnet · Claude Mythos Preview · Claude Opus 4 (x) · Command R+ · DeepSeek R1 · DeepSeek V3 · DeepSeek V3.2 · DeepSeek V4-Pro · Gemini 2.x · Gemma 2 · GPT-4 · GPT-5 · GPT-5.5 · Grok 3 · Llama 3 · Llama 4 · Mistral Large · OpenAI o1 · OpenAI o3