Anthropic CEO warns of AI-driven botnet 'swarm' taking over the entire internet — 'In 6–12 months such a swarm could be capable of taking over the entire internet with a persistent botnet'

Dario Amodei, CEO of Anthropic, has expressed concerns about the potential for an AI-powered botnet swarm to take over the entire internet within 6-12 months. The rapid advancement of AI technology has led to intense competition, with four American companies currently developing frontier AI models. However, Amodei warns that the lack of necessary guardrails could result in catastrophic damage, potentially exceeding hundreds of billions of dollars.

AI scientist Evan Hubinger, a former Anthropic employee, has shared a similar perspective, suggesting a 10% chance of human extinction by the end of the decade due to AI. In an X post, Hubinger emphasized the severity of the risk, stating that "we really do earnestly believe AI could kill all humans."

While some see AI as a potential threat, others believe that increased capabilities could also lead to stronger safeguards and automation of non-destructive tasks. Nevertheless, the development of AI without proper safeguards poses significant risks. AI models can now assist with weapons engineering, military intelligence, surveillance, and cyber operations, among other destructive activities. More advanced systems could drastically reduce the expertise, manpower, and time required to carry out these tasks.

The potential dangers of AI have been likened to those depicted in science fiction, such as James Cameron's Terminator and Frank Herbert's Dune. These stories warn of the consequences of losing control over autonomous military AI and becoming dangerously dependent on intelligent systems. However, the key lies in ensuring that AI systems are developed with appropriate controls and safeguards in place to prevent misuse and mitigate potential harm.