AI-linked stocks fall after call for development slowdown worries investors – business live
Rolling coverage of the latest economic and financial news AI CEOs say they need to slow the pace of development. But will they? The Guardian view on controlling AI: humanity cannot outsource its survival It is “self-serving” for AI companies such as OpenAI and Anthropic to talk of slowing things down, argues investor Michael Burry. Posting on X, Burry ( famous from The Big Short ) suggests four…
AI-linked stocks experienced a decline following a call for development slowdown, causing concern among investors. AI CEOs have suggested the need to reduce the pace of development, but investor Michael Burry argues that this stance is self-serving. According to Burry, there are several reasons why AI companies like OpenAI and Anthropic may benefit from promoting a slowdown:
1. LLMs (large language models) are not AI and will not lead to the creation of AGI (artificial general intelligence), so there is nothing to slow down.
2. The competition in the AI sector is heating up quickly, so slowing down benefits incumbents.
3. IPOs (initial public offerings) rely heavily on hype and puffery, and discussing a slowdown could be seen as hype and puffery.
4. Real, uncontrollable slowing growth could be used to justify a push out of IPOs, as investors may be pushed out.
The competitive race between companies and countries in the AI field is intense, and it appears challenging for firms to voluntarily step back while rivals continue to push ahead. President Trump's recent comments on the matter further emphasize the unwillingness to slow down, even if it may benefit safety, monitoring, and governance of increasingly powerful systems.
Some argue that the debate may alter the composition of AI capex (capital expenditure) more than its overall scale, directing a greater share of investment towards safety, monitoring, and governance alongside the continued build-out of compute infrastructure.
Written by urgent.news from Guardian Business's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
