AI agents are scrambling power users' brains (www.axios.com)

Recent discussions in the tech community have highlighted a concerning trend among developers using agentic AI coding tools like Anthropic's Claude Code and OpenAI’s Codex, which can autonomously write and ship software. Prominent figures, including OpenAI co-founder Andrej Karpathy, have reported experiencing symptoms akin to addiction, with some coding for up to 16 hours straight while urging themselves to maximize their token use before month's end. This shift is raising alarms about "AI psychosis," where the boundaries of healthy work-life balance are blurred, leading developers to sacrifice sleep and mental well-being in pursuit of productivity. Experts warn that the reliance on these tools can induce a state of "brain fry," a term coined to describe the cognitive overload resulting from excessive AI tool oversight. The addictive nature of these technologies, likened to gambling, amplifies this issue, particularly among those in high-pressure tech environments. Carnegie Mellon’s Tim Dettmers emphasizes that while AI agents increase productivity, they also exacerbate the struggle with focus and mental bandwidth, potentially leading to burnout and higher employee turnover. As more developers integrate these tools into their workflow, the implications for mental health and work culture in the tech industry could be profound.