OpenAI and Anthropic changed customer terms six times in one week; here's how to bulletproof your workflow
brucemacv · x · 2026-11-03
mardehaym notes OpenAI and Anthropic changed rules on customers six times in one week: OpenAI said it would halve Codex usage on its $200 plan, then reversed and doubled limits instead, put ChatGPT and Codex on one shared usage meter, and capped paid accounts per device; Anthropic renamed top plans, changed how limits are counted, and made Claude Code refuse to rebuild existing video games. Any of these can break live workflows or spike bills overnight.
brucemacv relays four defensive measures:
- Real test cases you re-run whenever a vendor changes anything
- One gateway for every model call, so switching vendors is a config change
- One usage report per workflow, so finance sees limit changes as cost lines
- Keep your rules and data in your own systems so you can move them
More from coding & agent
- Obsidian dev open-sources Pele, a Mermaid-compatible renderer 100-500x faster than mermaid.js — Teknium · 2026-11-03
- Agent-built eval harness spawns subagents via codemode in one durable session — mitsuhiko · 2026-11-03
- Impeccable crosses 100k GitHub stars, with weeks of releases teased ahead — jamdac · 2026-11-03
- Every to demo 'Compound Sandwich': which engineering workflows to hand to agents — every · 2026-11-03
- Every launches an AI coworker in Slack that turns tagged tasks into reusable team skills — every · 2026-11-03
- Founder runs 2 AI companies with 8 Grok bots: Chief, PA, email, ads and GTM agents — alexgoughcooper · 2026-11-03