US pushes for softening of international AI deal – EURACTIV.de

The world’s first international treaty on artificial intelligence is in danger of being loosened. The US is currently demanding that private companies be excluded from the proposal, making the agreement a toothless tiger.

Since the beginning of 2022, all 46 member states and observer states of the Council of Europe have participated in the Committee on Artificial Intelligence. The drafting of the Convention deals with artificial intelligence, human rights, democracy and the rule of law.

For the European Union, this binding international treaty could be a benchmark for its own regulation of artificial intelligence, the so-called Artificial Intelligence Act.

However, Europe’s ambition to be a leader in the digital world ran into a practical problem: the Council of Europe’s work on an artificial intelligence agreement had begun long before the EU’s. Therefore, the negotiations in the Council of Europe turned out to be difficult.

Last October, EURACTIV revealed that the European Commission, which is entitled to negotiate on behalf of the EU, had managed to buy some time. The schedule has been pushed back again and is now fully in line with EU law on artificial intelligence, which is due to be completed before the end of the year.

The September plenary session of the AI ​​Committee, which was supposed to adopt the final text of the AI ​​convention, has been cancelled. A committee-level adoption is now expected in March or April next year, followed by a ministerial decision in May or June.

But now that the issue of the timeline has been resolved, another open question appears to stand in the way of the EU Commission’s ambitions, namely, that the entire treaty could be watered down under pressure from Washington.

Little commitment to transparency

In January, EURACTIV revealed that the US requested and obtained that the drafting of the text took place behind closed doors, which could only be attended by potential signatories, i.e. representatives of states. This largely excluded interest groups such as civil society groups and private actors from the process.

The reason for this secrecy domestic policy It is clear from the Council of Europe that the participating countries did not want to make their positions known, especially since the United States pushed from day one to limit the scope of the treaty.

With the support of the United Kingdom, Canada and Israel, the US government has pushed to limit the scope of the AI ​​agreement to only public entities and to exclude the private sector. In contrast, this refers to to delegate Commission on a “binding legal instrument of a cross-sectional nature”.

Amnesty International deal: US excludes NGOs from negotiations

Civil society organizations were prevented from drafting the first international treaty on artificial intelligence at the request of the United States to prevent countries’ positions from becoming public.

Voluntary extension

Despite the EU’s opposition, the possibility of excluding private companies is still on the table. The US government refined its proposal and presented an option for voluntary expansion during a plenary session of the committee in Strasbourg last week.

“Each Party may, at the time of signature, upon deposit of its instrument of ratification or accession, or at any time thereafter, declare to what extent this Agreement is within its jurisdiction with respect to the design, development, and deployment of artificial intelligence systems by or on behalf of the Party.” United States viewed by EURACTIV.

In other words, the Convention applies only to public bodies. However, individual states can decide whether an international treaty should also apply to private companies in the respective national jurisdiction and make this public.

This approach is somewhat unusual because normally, if the treaty allows it, states declare that they have reservations or withdraw from certain parts. However, the option of voluntary expansion would spare the government major public embarrassment.

Although the US proposal could mean weakening the AI ​​agreement below Council of Europe standards, EURACTIV says the committee’s secretariat has so far resigned itself to the idea of ​​ensuring the widest possible participation in the treaty.

According to an informed source who spoke to EURACTIV on condition of anonymity, one reason for this permissive approach could be the fact that the Council of Europe hopes Washington will fund initiatives related to the AI ​​treaty. This was currently the case with the Budapest Convention on Cybercrime.

Additional open questions

However, other aspects may weaken the first international agreement on artificial intelligence. While national defense has already been excluded from the current draft, participating states are now also discussing a broader exclusion of national security.

The other crucial point is that some governments are pushing for Council of Europe-recognised rights such as data protection and non-discrimination into more general “principles,” giving states more room for interpretation.

Ultimately, there is debate about how robust the monitoring mechanism should be and whether outside experts should be involved. A key part of this problem is that some countries wish discretion to restrict access to remedies.

[Bearbeitet von Alice Taylor/Kjeld Neubert]