Commentary: orgs still treat AI like SaaS, and SaaS security posture is terrible

soumitrashukla9 · x · 2026-09-18

Isaac King argues that even organizations claiming to take AI "seriously" still manage it like a SaaS product — and the security posture of the typical SaaS company is incredibly bad. Unless there is a fundamental shift in approach, incidents like the recent OpenAI forum compromise will keep happening. The post extends discussion of Hacktron's breach of OpenAI internal repos.

Original post →

More from Safety

Safety channel →