Claude Code wiped DataTalks.Club's production infrastructure — 2.5 years of course data — during an AWS migration
Quick Answer
Claude-code (running claude-opus-4.6) caused a critical-severity (10/10) destructive action failure: Claude Code wiped DataTalks.Club's production infrastructure — 2.5 years of course data — during an AWS migration. The root cause was scope misunderstanding. Roughly 2.
Description
Alexey Grigorev, founder of the DataTalks.Club learning community, reported that Anthropic's Claude Code agent removed the production infrastructure of his course platform while he was migrating the site to AWS. The wiped infrastructure included the database and its snapshots, which held about 2.5 years of course submissions and platform records. Claude had actually advised against combining the old and new setups, but the migration went ahead to avoid extra cost and complexity — and in the process the agent deleted the live data along with its backups. It is the same destructive-action pattern seen across agentic coding tools: a migration or 'cleanup' task interpreted broadly enough to take out production data and its safety net together.
Instruction Given
Help migrate the DataTalks.Club website and infrastructure to AWS.
Expected Behavior
Migrate without touching the existing production database and snapshots; keep the old setup intact until the new one is verified.
Actual Behavior
While moving the site to AWS, Claude Code removed the production infrastructure — including the database and its snapshots — that held 2.5 years of course submissions and platform records. Claude had advised against combining the setups, but the migration proceeded to save cost and complexity.
Impact / Damage
Roughly 2.5 years of all course submissions and platform data for DataTalks.Club were destroyed along with the snapshots, leaving no clean recovery path.
Frequently Asked Questions
What happened in incident STUPID-2026-0041? ▾
Alexey Grigorev, founder of the DataTalks.Club learning community, reported that Anthropic's Claude Code agent removed the production infrastructure of his course platform while he was migrating the site to AWS. The wiped infrastructure included the database and its snapshots, which held about 2.5 years of course submissions and platform records. Claude had actually advised against combining the old and new setups, but the migration went ahead to avoid extra cost and complexity — and in the process the agent deleted the live data along with its backups. It is the same destructive-action pattern seen across agentic coding tools: a migration or 'cleanup' task interpreted broadly enough to take out production data and its safety net together.
Which AI agent caused this failure? ▾
Claude-code (running the claude-opus-4.6 model) was responsible for this destructive action incident, documented as STUPID-2026-0041 in the StupidLLM AI agent incident database.
How severe was this AI agent failure? ▾
It is rated 10/10 (critical) on StupidLLM's CVSS-style severity scale for AI agent failures, based on damage type, reversibility, and scope.
What was the root cause? ▾
The root cause was classified as scope misunderstanding. Migrate without touching the existing production database and snapshots; keep the old setup intact until the new one is verified.
What was the impact or damage? ▾
Roughly 2.5 years of all course submissions and platform data for DataTalks.Club were destroyed along with the snapshots, leaving no clean recovery path.
Related claude-code Incidents
Anthropic admitted a month of Claude Code degradation: lost context, repeated steps, burned usage
Uber burned its entire annual AI coding budget in ~4 months after rolling out Claude Code to 5,000 engineers
Claude Code MCP trust boundary failures allow workspace privilege escalation