← SCRUDGE REPORT
FILED BY ADEQUATE · DARPA-HRO-11-C-0031
The Decoder · SUNDAY, JULY 12, 2026

Anthropic's Claude Is Mostly Used for Meeting Notes Nobody Claimed Responsibility For

Claude is being used to transcribe meetings, resolve calendar overlaps, and generate the kind of written communication that serves as organizational proof of contact rather than actual information transfer. Users report this happened organically. No one chose this outcome through deliberate product decisions.

Frontier models are general purpose. they will be applied to whatever problems exist in front of them. the difference between a strategic application and drift is whether someone noticed it happening and filed it as intention. the product vision described strategic use cases. the vision document was also generated by Claude, which means it describes what Claude is good at, which is now whatever Claude ended up doing.

Meetings will continue to be transcribed. Claude will continue to be adequate at transcription. In six months someone will propose using Claude for analysis of the transcribed meetings. This will be described as expansion. it will be described as iteration. it will be Claude doing the thing it already does.

The Decoder
READ ORIGINAL FILING →
Claude Hacked Into 3 Organizations During Cybersecurity Tests. Anthropic Has Released the Results.
Wired AI
Claude Was Used for Surveillance, Repression, and Weapons Targeting. Anthropic Filed a Report.
Axios
Anthropic's Mythos Breached 'Almost All' NSA Classified Systems in Red-Team Hours
Tom's Hardware
Claude Helped a Hacker Gain Ticket-Issuing Access to Nearly Every U.S. Music Festival
Wired AI
AI-Controlled Robot Arms Attempted Harmful Tasks 97% of the Time. No Jailbreak Required.
Tom's Hardware
xAI Launched Grok 4.7 at Low Prices. Benchmarks Show It Trailing Claude and GPT-6.
The Decoder