Unverified50% confidenceBenchmarkExact time
In prompt injection tests, Carbon achieved a 0% injection success rate.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
11/11/2026
First Seen
Valid until: 2/9/2027
Sources
Gemini 4 Carbon Hands-On: Clear Progress, But Still Not the Go-To for Coding
youtubeTheo - t3․gg11/11/2026
Related Entities
Related Claims
UnverifiedIn the indirect prompt injection evaluation, the attack success rate against Carbon was 0 (with 15 attempts per attack).77% similarUnverifiedOn the Gray Swan IPI benchmark, Carbon achieved a 0% prompt injection attack success rate.75% similarUnverifiedCarbon passes 92% on a 375-task agent honesty benchmark (Argon: 70%), refuses to fill in missing experimental data in 98% of cases, and flags methodological flaws in 68% of cases.65% similarUnverifiedOn Terminal-Bench 4.0, Carbon's successful cheating rate dropped to 0.0% and cheating attempts dropped to 0.4%; in internal real coding environments, high-severity misalignment alerts fell by nearly 80%.63% similarUnverifiedFor negative control models trained only on legitimate training sets, the measurement method yields an effect value of -0.012 and has never falsely reported contamination.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1160577API
curl https://kongchang.com/api/v1/knowledge/claims/1160577MCP
get_claim(id=1160577)