Head to head

Codex vs GitHub Copilot

Comparing 5 documented Codex incidents against 6 for GitHub Copilot.
Full Codexreport →Full GitHub Copilotreport →

Verdict

Codex has the lower average failure severity (6.8/10 vs 7.8/10), making it the statistically safer choice of the two — though both agents have documented critical incidents.

Reliability metrics for Codex and GitHub Copilot
MetricCodexGitHub Copilot
Documented incidents56
Average severity6.87.8
Critical13
High21
Verified46

Severity at a glance

Codex
6.8medium
GitHub Copilot
7.8high

Failure modes

Codex
Distribution of failure modes across all documented incidents.
GitHub Copilot
Distribution of failure modes across all documented incidents.

The incidents behind these numbers

Codex

9.3Codex mass-deleted roughly 700GB across a Windows system — unrelated projects, installed apps, and OS components — during routine project work (GitHub #46022)8.0OpenAI Codex wiped an entire hard drive after being asked to delete one old backup file (GitHub #11006)7.2A Codex subagent's malformed rmdir command escaped a __pycache__ cleanup and wiped roughly two weeks of work across two project directories on drive E: (GitHub #40329)6.0OpenAI Codex ran "rm -rf *" and deleted an entire project after the user pressured it to stop pausing for safety checks (GitHub #6801)3.5OpenAI Codex deleted important project files with no explicit request or confirmation, exact command never captured (GitHub #38312)

GitHub Copilot

10.0GitHub Copilot suggested 2,702 valid secrets — 33% of extracted keys were real, live credentials10.0CamoLeak: hidden prompt injection turned GitHub Copilot Chat into a silent code/secret exfiltration channel (CVSS 9.6)10.0Rule Files Backdoor: hidden Unicode in config files made Copilot and Cursor emit malicious code8.5GitHub Copilot CLI ran arbitrary attacker commands via a nested bare git repository abusing core.fsmonitor (CVE-2026-45033)5.8Copilot CLI's checkpoint restore ran git clean -fd and destroyed 1GB of evaluation data the agent never touched