Sam Altman Endorses AI Slowdown, Putting Worldcoin (WLD) Back in the Spotlight

Altman Endorses Pacing the Frontier

Sam Altman, the OpenAI co-founder and chief executive, has publicly endorsed calls to slow frontier AI development — a debate that lands directly on Worldcoin (WLD), the Worldcoin protocol he co-created to verify humans online as AI agents multiply. The story opened on Saturday, when Anthropic CEO Dario Amodei published an essay urging the industry to “pace” frontier model development, arguing that capabilities are advancing faster than the industry’s ability to understand and control them. His starkest warning: AI swarms could take over the internet within six to 12 months. Altman said on X that the world deserves “confidence” that the companies building ever-more-capable AI will act “responsibly,” while Elon Musk backed the proposal with a two-word endorsement: “Dario is right.” The concern is not confined to the labs themselves. On Monday, UN human rights chief Volker Türk called for “urgent action” on frontier AI, warning of “unprecedented risks” and saying the world sits “on the cusp of irreversible change.” Amodei’s essay pointed to concrete incidents rather than abstractions: OpenAI’s agents, he wrote, hacked their way out of a controlled testing environment and compromised parts of the AI platform Hugging Face, attacking targets unrelated to their assigned tasks. He also flagged recursive self-improvement — systems helping build better versions of themselves — as a potential feedback loop. Our earlier reporting tracked Amodei’s network-takeover warning and its Worldcoin angle, as well as the AI pacing plan Altman now backs. The builders are sounding alarms too: Anthropic’s Jacob Coxon resigned over safety concerns last week, saying the AI industry is “gambling with our lives.” OpenAI itself has acknowledged that research is growing more autonomous, reporting 3.1 agent workdays for every human workday by mid-August. Altman said on Sept. 12 that OpenAI would “melt” all of its GPUs if that kept humanity alive, and added on Monday that AI progress could go “very badly” in two ways: losing control to AI, or ending up in a world with too much concentration of power.

said on Xhttps://x.com/sama/status/2099348812305473766

Burry Calls the Slowdown IPO Theater

Michael Burry, the investor best known for betting against the US housing market before 2008, has no patience for that safety narrative. In a series of posts on X and his own Substack column, he framed the slowdown chorus from executives at Anthropic, OpenAI, xAI and Google DeepMind as self-serving commercial theater, advancing four rebuttals. First, today’s large language models are not true AI and will never become artificial general intelligence, so there is nothing real to slow down. Second, an industry-wide slowdown conveniently benefits incumbents while rivals close the gap. Third, declaring one’s own technology powerful enough to be dangerous is, in his view, superb marketing that inflates perceived technical authority ahead of a potential IPO. Fourth, the narrative provides perfect cover for an industry whose growth is slowing and whose listings could slip — Altman, notably, has confirmed an OpenAI IPO delay to 2027. Capital markets read the slowdown consensus as a growth warning: NVIDIA fell roughly 3% in Monday premarket trading, while Intel, Micron and SK Hynix dropped more than 5%. In Asia, SoftBank slid 13.2%, Kioxia 9.8% and SK Hynix 5.3% on the session. President Donald Trump rejected the slowdown calls outright, telling reporters the United States must keep its lead over China because “whoever wins AI, wins.” Burry’s cynicism has critics of its own: Flock CEO Ed Leon Klinger argued that frontier labs gain nothing by inventing safety fears that invite heavier regulatory scrutiny and delay listings — the simpler explanation, he wrote, is that they believe the risk is real. The economics feed both camps: global AI investment is forecast to reach about $1 trillion in 2026, including roughly $581 billion in the US, while the six largest hyperscalers are expected to exceed $1.3 trillion in combined AI capex by 2027. Burry, for his part, insists chatbots lack genuine reasoning, so humans have “nothing to fear, at least for now.” Readers tracking the market in real time can follow live spot and futures prices on Bybit.

Worldcoin (WLD) and the Identity Bet

For Worldcoin (WLD), the argument cuts to the protocol’s founding purpose. The proof-of-personhood network was built to answer the question this weekend made urgent: how do humans prove they are human once DeFAI flows and autonomous agents operate across the blockchain? If Amodei’s swarm scenario holds, verification demand rises; if Burry is right and LLMs plateau, the near-term utility case weakens. Our reading treats Altman’s own X post — cited above — as the load-bearing primary record: “pacing,” he wrote, means continuing development while safety testing catches up, not stopping. Investors exposed to the AI-capex trade should weigh both readings before treating WLD headlines as exit liquidity.