Anthropic CEO warns of rogue AI agent swarms taking over internet

anthropic ceo dario amodei recently warned in an essay that swarms of rogue ai agents could potentially take over the entire internet in as little as six months. however, experts gary marcus, nathan hamiel, and zack korman strongly disagree with this assessment, calling it vague and implausible. they argue that taking down major internet infrastructure, such as cloudflare, google, and aws, would be extremely difficult, if not impossible, due to the companies' robust security measures and the complexity of coordinating such an attack. additionally, the experts question the motive behind such an attack, as it would likely cause significant harm to the attackers themselves and would require substantial resources. this debate highlights the ongoing concerns and uncertainties about ai safety, regulation, and the potential risks associated with advanced ai systems.