Unverified50% confidenceBenchmarkExact time
Gemini 4 Argon's failure rate on code audit benchmarks ranges between 50% and 70%, with all three independent projects showing high false-positive rates.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/14/2026
First Seen
Valid until: 1/12/2027
Sources
Related Entities
Related Claims
UnverifiedGemini 4 Argon scores 15% on the hallucination benchmark (lower is better), lower than any current frontier model.78% similarUnverifiedOn a test set of 100 tickets, SupportOps-AI had an escalation error rate of 18%, a hallucination rate of 6%, an infinite-loop rate of 3%, and a complete failure/unresolved rate of 4%.71% similarUnverifiedIn the test fixture, Atlas is the only lead automatically routed to sales (confidence ≥ 0.9 and score ≥ 80); Birch enters manual review; Delta has a confidence of 0.83 and enters manual review even with a high score; Cinder falls below 0.65 and enters the nurturing process.67% similarUnverifiedIn Artificial Analysis's AA Omniscience test, Argon has a hallucination rate of 15% (the lowest among models with an intelligence index of 45 or above), and an accuracy rate of 50%, lower than Astra's 63%; Argon tends to admit it does not know rather than offering an incorrect guess.66% similarUnverifiedIn the control test, relying solely on prompts to instruct the AI to write less code saved about 30% of lines, but security dropped to 95%.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1030749API
curl https://kongchang.com/api/v1/knowledge/claims/1030749MCP
get_claim(id=1030749)