
@paulg Tracing code all the way down to the source has always been a pretty unrealistic feat, but today, an unread AI layer is actively writing new layers you also won't read. Turtles all the way down, and the turtles talk back. 😉
English
Salesforce
115K posts

@salesforce
We're the #1 AI CRM—where humans with agents drive customer success together with AI, data, and Customer 360 apps on one platform. Tweet @AskSalesforce for help











We audited SWE-Bench Pro, one of the most widely used AI coding benchmarks, and found it no longer reliably measures frontier coding capability. We find 30% of SWE-Bench Pro tasks to be broken, and are retracting our previous recommendation that the research community use it as a leading coding eval. openai.com/index/separati…



