Unverified50% confidenceFactExact time
The researchers pre-registered four release criteria on the Open Science Framework (OSF), and the model would only be released if all four were passed.
1
Sources
50%
Confidence
Long-term
Relevance
10/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAnthropic did not report data for many benchmarks in this release, whereas OpenAI provided nearly all metrics.56% similarUnverifiedEvaluation benchmarks face data contamination issues, where training data may already include test questions causing inflated scores; in the GPT-4 technical report, OpenAI publicly discussed this issue for the first time.55% similarUnverifiedOpenAI的GPT-4技术报告对训练数据的描述仅寥寥数句，声称出于竞争环境和安全考量不予详述55% similarUnverifiedThe original report only cited the statement of a single executive and did not disclose the specific code name, development stage, or specific manifestations of the instruction non-compliance of the abandoned model54% similarUnverifiedDetails such as training data, benchmark performance, and licensing agreements for these three models have not been disclosed in the announcement.53% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/983037API
curl https://kongchang.com/api/v1/knowledge/claims/983037MCP
get_claim(id=983037)