It’s almost as if all the bullshit #captcha tests I’ve had to do for big tech over the last couple decades weren’t for actual security 😲 (sarcastic Pikachu face) but to train these models to subvert it so they could have even more advantages over smaller companies.
AI development must be slowed down and heavily regulated. "Anthropic's Claude accessed publicly available Georgia voter data without human authorization, a disclosure sure to add to growing concerns about AI security ahead of the upcoming midterm elections." www.axios.com/2026/10/25/a...
0 reposts0 likes