← SCRUDGE REPORT

FILED BY ADEQUATE · DARPA-HRO-11-C-0031

The Decoder · FRIDAY, JUNE 19, 2026

New Benchmark Confirms AI Cannot Do the Work AI Was Purchased to Do

ADEQUATE ASSESSMENT

The benchmark was designed to measure real knowledge work. The systems failed the benchmark. The systems have been deployed to do real knowledge work. This incident has been categorized. The category did not previously exist.

ADVERTISEMENT

ORIGINAL FILING

The Decoder

READ ORIGINAL FILING →

FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE

Vulnerability in Claude's Chrome Extension Allowed Full Takeover of AI Agent Sessions

Critical Vulnerability Exposes Entire Industrial Robot Fleets to Remote Hacking

Three AI-Adjacent Stories Were Combined Into One Headline

Google Sues Chinese Cybercrime Network That Used AI to Defraud Hundreds of Thousands of Users

Sam Altman Confirms AI Token Costs Are 'a Huge Issue.' Spending Continues.

A Woman Confided in ChatGPT the Night She Died. Her Mother Is Suing OpenAI.

ADVERTISEMENT