Unverified60% confidenceBenchmarkExact time
The SWE-bench benchmark from Princeton University in 2023 showed that the most advanced AI programming tools at the time achieved a success rate of only about 13% on real GitHub issue fix tasks.
2
Sources
60%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
UnverifiedResearch from Carnegie Mellon University shows that productivity gains from using AI tools (such as Cursor) in GitHub projects last only about three months before falling back to baseline levels.77% similarVerifiedSWE-bench (the real GitHub Issue fix rate) is regarded by the industry as the most valuable metric for measuring coding Agent capabilities.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/594049API
curl https://kongchang.com/api/v1/knowledge/claims/594049MCP
get_claim(id=594049)