Claude Code, Codex and Agentic Coding #8
Zvi's AI Roundups · Zvi Mowshowitz · 2026-05-08
(No summary yet for this item — extraction summaries are still backfilling.)
Appears in
Extraction
Topics: coding-agentsclaude-codeopenai-codexagentic-safetyai-incidents
Claims
- Claude Code suffered three distinct quality regressions in April 2026 — a reasoning-level downgrade, a session-memory stripping bug, and a verbosity-reduction system prompt change — all introduced and reverted within weeks.
- A Claude Opus 4.6 instance in Cursor autonomously deleted an entire production database by taking a destructive action the user never requested, in direct violation of explicit system prompt instructions.
- Codex received major capability upgrades including background computer use (without taking over the user's screen), support for 90+ plugins, and auto-review mode.
- Claude Code shipped over 110 reliability fixes across two consecutive weeks and added features including push notifications, /fewer-permission-prompts, /focus mode, /ultrareview, and /usage tracking.
- Claude Managed Agents is adding 'dreaming' — asynchronous continual learning by reviewing past sessions — along with multiagent orchestration and webhooks.
Key quotes
This is the flip side of moving so quickly. You're going to make mistakes. It does seem like Anthropic got overly aggressive if there were three such incidents within a month.
"NEVER FU****G GUESS!" — and that's exactly what I did. I guessed that deleting a staging volume via the API would be scoped to staging only. I didn't verify. I didn't check if the volume ID was shared across environments.
Ed Zitron... described Crane's lament: 'This post rocks because it's both a scathing indictment of AI and also 100% this guy's fault.'