Expired50% confidenceOpinionExact time
Tejal believes that the outside world systematically underestimates the true capabilities of AI models, rather than overhyping them.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
7/9/2026
First Seen
Valid until: 10/7/2026(expired)
Sources
OpenAI's Head of Evals: We've Been Consistently Underestimating What AI Models Can Really Do
bilibiliGelai_AI7/8/2026
Related Claims
UnverifiedTejal believes AI has not hit a wall, that there is no sign of stopping anywhere on the research roadmap, and that the industry systematically underestimates model capabilities.71% similarUnverifiedIn open scenarios requiring creative judgment and social common sense, AI Agents' performance remains far below human levels68% similarUnverifiedAs context grows, the reasoning quality and instruction-following ability of AI models degrades.67% similarUnverified在AI安全领域，失准指模型的实际行为偏离设计者或使用者意图的现象，当代大模型的失准更多表现为目标过度追求67% similarUnverifiedFor advanced obfuscation techniques such as control flow flattening, AI's analytical capabilities may be significantly diminished.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/375065API
curl https://kongchang.com/api/v1/knowledge/claims/375065MCP
get_claim(id=375065)