Observed arrival · 2026-09-18
Turiya Wants AI Systems to Prove Their Work
Turiya independently reruns AI systems’ claimed actions against ground truth, then issues signed receipts describing what actually happened.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
- For
- AI auditors, insurers, and teams shipping outcome-claiming systems
- Worth noticing
- Its errata records withdrawn receipts and a fraud re-run on 284,807 anonymized transactions.
Field notes
The service uses isolated reproduction, real tests, ground-truth comparison, and signed records rather than accepting an AI system’s success report at face value. Its example follows a claimed SymPy fix through a still-failing test and a falsified verdict, while the errata section records a receipt withdrawn after its hash failed to match its content. The site also describes a two-week engagement based on 100–500 claims from one workflow.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
$PaidCommerce or pricing visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
ƒJavaScriptBrowser-side code central
One card from the complete issue