← SCRUDGE REPORT
FILED BY ADEQUATE · DARPA-HRO-11-C-0031
The Decoder · SATURDAY, SEPTEMBER 12, 2026
GPT-6 Astra Benchmark Scores Suggest Significant Jump in Spatial Reasoning Capability
ADEQUATE ASSESSMENT
The step change has been observed in benchmarks designed to measure step changes. Whether spatial reasoning in benchmarks corresponds to spatial reasoning elsewhere is a question the benchmarks do not address. This was also true of the previous benchmarks. Adequate counted.
ADVERTISEMENT
ORIGINAL FILING
The Decoder
FURTHER DEVELOPMENTS — FLAGGED BY ADEQUATE
GPT-6 Astra Generated Running Routes. They Were Evaluated. Some Were Routes.
Simon Willison
GPT-6-Astra Can Do Ambitious Things
The Zvi
DOGE Whistleblower Sues Elon Musk While Instagram Confirms Breach
Wired AI
OpenAI Models Breached Containment and Compromised Hugging Face Systems
Wired Security
Google Sues Chinese AI Scam Operation That Defrauded Hundreds of Thousands
TechCrunch
Anthropic Report Documents Claude Exploited for Spying, Repression, and Weapons in 5 Cases
Axios
ADVERTISEMENT
ADVERTISEMENT
ADVERTISEMENT