Canadian PM Urges AI Diversification After US Anthropic Block, Decentralized AI Tokens Rally – Decrypt

Canadian Prime Minister Mark Carney warned Sunday that U.S. restrictions on Anthropic’s most superior AI fashions expose the danger of leaning on a handful of American suppliers, after a authorities order pulled two frontier programs offline for customers worldwide.

“The state of affairs we’re in collectively proper now with Mythos and Fable is one thing that may occur with overreliance on sure fashions,” Carney mentioned, talking in Eire forward of the G7 summit in France, the Related Press reported.

Carney mentioned no person has “achieved something incorrect within the state of affairs,” however warned the error could be to “simply settle for this, do not take the lesson, do not construct out and diversify.”

“It’s by no means a good suggestion to have one choice,” the prime minister added.

Carney’s remarks observe a directive on Friday ordering Anthropic to chop entry to its Fable 5 and Mythos 5 fashions for any overseas nationwide inside or outdoors america on nationwide safety grounds.

Anthropic promptly complied, disabling each programs for all prospects. It disputed the premise, nonetheless, arguing the cited jailbreak is already replicable on public fashions like OpenAI’s GPT-5.5.

Commerce Secretary Howard Lutnick reportedly despatched the letter to Anthropic CEO Dario Amodei, per Axios. It was reportedly pushed partially by suspicion {that a} China-linked group had accessed Mythos, Semafor reported.

The disputes come as Anthropic nears a $1 trillion valuation following annualized income of over $47 billion.

Decrypt has reached out to Anthropic for remark and can replace this text ought to they reply.

Dangers and prices

The prime minister’s issues solid centralized AI as a single level of failure, the place one firm’s compliance with a authorities order can lower off customers in every single place directly.

Centralized AI retains a mannequin and its controls inside one firm. Decentralized AI spreads these features throughout impartial operators coordinated by a blockchain.

Tasks tied to decentralized AI rallied following the ban on Anthropic’s fashions, with the sector’s market cap at $24.3 billion, up 6% on the day and 12% over the week, per information from CoinGecko.

Smaller compute and information networks led the best way, with ChainOpera AI, io.web, Grass, and NOVA amongst these climbing greater than 30% over the previous week, whereas NEAR Protocol and Bittensor, two of the sector’s largest tokens by market cap, rose 15.9% and 27.9% on the week respectively.

The U.S. authorities’s transfer on Anthropic factors to “a danger that’s largely distinctive to centralized AI,” Dan Dadybayo, technique lead at Horizontal Programs, informed Decrypt. Backing Carney’s warning, Dadybayo mentioned reliance on a number of U.S. suppliers “creates real systemic danger, just like what we noticed in finance in 2008.”

Distributing fashions throughout impartial nodes removes the only “kill change,” however the danger persists if the compute behind them stays concentrated amongst a number of suppliers, Dadybayo defined.

Such a pattern is being pushed by “rising compute and information prices,” Peter Anthony, founder and CEO of Perceptron Community, informed Decrypt. Anthropic’s state of affairs “did not create that downside, it simply made it unimaginable to look away from,” he added.

Whereas Anthony agreed with Carney’s level that the order marked a “strategic vulnerability,” he questioned whether or not decentralization solves it or “simply pushes the chokepoint again one layer to GPU suppliers.”

If decentralized AI nonetheless runs on chips owned by a number of cloud giants, “you have rebranded the danger, not eliminated it,” he added.