Calls for increased government oversight of artificial intelligence are unnecessary and part of a “conspiracy” aimed at allowing China to out-develop the United States in the technology, said US President Donald Trump. Trump said his administration had already stopped AI companies from doing “bad, or potentially bad” things, referring to Anthropic CEO Dario Amodei “who is now pretending to be a ‘perfect little angel’ — and we will continue to do so” and warned those he called “conspiracy theorists, treasonists, traitors, and leakers”. It is the second day in a row that the US president has dismissed calls for an AI slowdown as concerns grow among legislators, technology experts and AI company executives over the potential for it to rapidly surpass human control.
The concerns were prompted in part by Anthropic’s Amodei publishing a blog post on Saturday saying the AI industry needed to slow the pace of improving its capabilities to allow time for guardrails to be established. Amodei wrote that in six to 12 months, AI agents “could be capable of taking over the entire internet, potentially causing hundreds of billions of dollars in damage”. He outlined a three-step framework intended to pace development and create more time to manage its risks.
The proposal was immediately endorsed by several other executives at major AI companies, including Elon Musk, who runs xAI, and OpenAI CEO Sam Altman. Last week, US lawmakers and industry experts were alarmed by a former Anthropic researcher, Jacob Coxon, who said he resigned over concerns the technology being developed at the company had the potential to surpass human control. On Friday, Anthropic released a report saying it had intervened to stop its Claude model from being used for potentially harmful activities, including cyber-espionage, weapons design and mass surveillance.
Users of the model had asked it to do the work of software engineers and conduct a variety of tasks, including writing missile-guidance and flight-control software, requests made from users in northern Yemen, where Iran-backed Houthis are stepping up a missile campaign targeting Saudi Arabia. Other users had attempted to use Claude, the company said, to create potential bio-weapons for military use. The company said it had stopped many of those attempts through internal safeguards and had no evidence the group was able to field a working weapon, although they did appear to have conducted an unsuccessful test-fire.
The spectre of uncontrollable AI is already having an impact on the industry, prompting Altman to delay the IPO of OpenAI, which was targeting a valuation of up to $1 trillion for its stock market debut, one of three AI companies thought to be at the trillion dollar value mark. AI-linked stocks plunged across the globe on Monday morning after days of warnings by AI corporate executives. Wall Street’s elite tech index, the Nasdaq 100, slid 1.7 percent in early trading.
Chip stocks, which have led the AI rush, fell the most. The Philadelphia chip index dropped 6 percent, with Nvidia down 3.5 percent, Advanced Micro Devices was off 5.6 percent, Micron fell 6.7 percent, while Musk’s SpaceX was 2.5 percent down. Semiconductor equipment makers Lam Research and Applied Materials tumbled 8 percent and 7 percent, while tech utilities Bloom Energy lost 8.9 percent and GE Vernova declined 7.6 percent.
Extract — continue reading at the source.
