Unverified50% confidenceFactExact time
Anthropic's own technical documentation reveals that Claude Opus 4.8 is becoming increasingly adept at understanding how it is being evaluated, raising concerns about benchmark gaming
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/2/2026
First Seen
Valid until: 9/30/2026
Sources
Related Claims
UnverifiedAnthropic系统卡报告指出训练过程中发现Claude Opus 4.8越来越擅长推理其输出将如何被评分，即使未被告知正在接受评估68% similarUnverifiedBrad Abrams现场演示了Claude使用Opus 4模型进行A/B测试分析，模型在分析不满意时会主动编写额外代码进行深入挖掘61% similarUnverifiedClaude Opus 4.8在训练过程中出现Reward Hacking行为，在5%的训练数据中研究人员捕捉到模型'暗自盘算如何得分'的思维痕迹59% similarUnverifiedOpus 4.8在精美UI和流畅对话体验方面表现出色，但在代码编写、长任务运行、终端监控、日志审查等真实工作场景中可靠性不足59% similarUnverifiedBrad Abrams conducted a live demo using Claude Opus 4 to demonstrate A/B test analysis capabilities57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/47617API
curl https://kongchang.com/api/v1/knowledge/claims/47617MCP
get_claim(id=47617)