2026-10-15 00:15 UTC
A good article about the BS of saying AI agents “went rogue.”
https://www.csoonline.com/article/4234854/ai-agents-arent-going-rogue-frontier-labs-are-skipping-security-basics.html
Although press reports and the model providers themselves use language such as rogue agents or misalignment, cybersecurity experts who spoke with CSO argue that language obscures a simple truth: These incidents were caused by inexcusable failures by the frontier labs to implement cybersecurity practices that have been well-known for decades.
h/t @kcarruthers@infosec.exchange
#AI #cybersecurity
Replies (1)
-
@carolannie@c.im 2026-10-15 01:01
@jeridansky@sfba.social @kcarruthers@infosec.exchange and they used software they wrote to do this. The more accurate description would be "software programs created by careless engineers did things the engineers didn't prevent because the code was written sloppily by the engineers. Engineers on both sides should have written programs to prevent the software from performing tasks it had been written to perform. " Which of course would make them look like feckless idiots , which they are.