AI coding agent leaders split by benchmark and workflow
Update, 14 September 2026: the vexp-swe-bench repository has been removed from GitHub and no archived copy exists, so none of…
-
-
DeepSeek Tests Open Model Economics; Foreign Coauthors
Today’s through-line is control. DeepSeek is reportedly trying to prove that open models and serious capital can coexist, while NIH,…
-
FutureSim Exposes Polymarket AI’s Narrow Wins and Failures
Max Planck Institute researchers recently released FutureSim, a benchmark for polymarket ai-style forecasting that tests whether agents can predict real-world…
-
Underwater Options Cut Real Pay to $7.5M: Xero CEO Pay
Xero disclosed on 14 May that Xero CEO pay in FY26 totalled $28.946 million on an accounting basis, but only…
-
$2 Billion on Paper: Sam Altman Self-Dealing and OpenAI
Sam Altman self-dealing claims entered the court record this week as Reuters reported that filings in the OpenAI case said…
-
AI Music Fraud is Soaring: Deezer’s 44% Warning
Deezer says AI music fraud now sits behind one of the biggest headline numbers in streaming: 44% of all new…
-
Data Center Backlash: Festus Voters Oust 4 Incumbents
Festus, Missouri just gave us a cleaner read on data center backlash than most national trend pieces do. A city…
-
Public Wealth Fund Means Citizens Still Ride AI’s Boom-Bust
A strange detail in OpenAI’s latest policy push is that the public wealth fund is being sold as a way…
-
GLM-5 vs Claude Opus: Why Cheap Models Win for Agents
YC‑Bench just produced the sort of result that usually launches a thousand hot takes: GLM‑5 vs Claude Opus on a…
