United States asks G20 not to introduce new artificial intelligence rules

The US position was presented at the G20 technology table, meeting in North Carolina. The White House’s technology adviser Michael Kratsios outlined a document called the “Carolina Principles,” whose content rests on two pillars: investment in fundamental research and limited government intervention. Washington is asking G20 countries not to introduce rules that could slow the development of artificial intelligence and opposes the creation of new international oversight bodies for the sector. The news comes from a Reuters dispatch picked up by AIdapted’s daily roundup: it currently has a single source, and no independent confirmation is available.

According to the same account, the American position fits within Washington’s attempt to maintain its technological edge over China. This is the point that makes the proposal less technical than it appears: an international oversight regime on artificial intelligence would not concern only the models, but also the timing with which they can be brought to market — and therefore the gap between those who arrive first and those who arrive later.

Two opposite directions on the same day

On the same day, two pieces of news showed what it actually means to move on artificial intelligence without waiting for a common framework.

The first comes from India, where autonomous payments managed by artificial intelligence were tested on the UPI infrastructure, the national digital payments system. Delegating a payment operation to an automated system is exactly the kind of application for which regulators usually demand prior safeguards: here the test comes before the rule.

The second comes from Helsinki. The city announced its intention to heat homes using waste heat produced by computing centers dedicated to artificial intelligence, feeding it into the city’s district heating network. This information is also reported by the same AIdapted roundup, based on Finnish sources, and remains single-sourced. The recovery of residual heat from computing centers is one of the technical responses to the growth in energy consumption linked to the expansion of artificial intelligence systems: instead of limiting the load, its waste is reused.

What distinguishes the three approaches

Taken together, the three elements describe different strategies on the same subject. Washington is asking for no additional constraints and for funding research upstream. New Delhi is testing a high-impact application on the public payments infrastructure. Helsinki is intervening on the physical side of the chain — energy and heat — without touching the question of rules on models.

None of the three paths implies the others, and none excludes the others. But the G20 table is the place where, should coordination emerge, it would have to pass through: the American request not to establish new international bodies concerns precisely that venue.

What we don’t know

We do not know the text of the “Carolina Principles” nor whether it was adopted, rejected, or merely presented: the available source reports its presentation, not its outcome. We do not know which G20 countries supported the American position and which opposed it. On the Indian test we do not know the scale, the operators involved, or the outcome; on the Helsinki project we do not have timelines, costs, planned thermal output, or the number of homes involved.

The only verifiable element with a date is the venue: the G20 technology meeting was held in North Carolina, and the American position was put on the record there.