The issue with these stories is AI brain people don't read them like normal people. This is a funny story for her not a total derision of everything she's working on. The inherent lack of safety or control is a a feature, computer man do funny thing is a selling point.
Yeah seems about right for this project. I really wanted this to be a serious browser, but nothing about this dude is serious.
Also I know he backed this statement up with much better testing but these AI brainrot things people say kill me: "I ran multiple passes of adversarial review, asking different models to analyze the code for mistakes and bad patterns."
But like what am I supposed to do when senior ai reporter Benj writes his next piece? Ars works because the writers are generally experienced in the topics and do analysis and provide insight. Do we just accept that chatgpt is the new head ai writer with a meat puppet? They need to address the trust issue before this is resolved.
I understand how agents work. I'm just saying it cannot "verify relevancy". That's a qualitative assessment that an LLM is incapable of doing. The scripts and regex that form the backbone of the "agent" can absolutely download a webpage and add the text contents to the context. But after that it's just random bullshit.
Be old/conservative, want to have sex at 18 because hormones, you can't unless you're married, get hitched for life to the first attractive person who shows up, wake up in 10 years and realize you married a stranger and you don't really get along.
One more gotcha in the AI booster arsenal: wrong model, wrong prompt, not enough agents, just don't look at it. While none of those things addresses the actual issues of watching everyone piss away their money into the pit for no reason other than psychosis.
Yeah but imagine reading about a new release of something and it appearing in your updates the same day. Shiny new software every day is addicting.