The European Commission is considering postponing some parts of the EU’s artificial intelligence (AI) rulebook as concerns over ethics surrounding training AI models and delays in developing comprehensive technical standards grow.
“If we see that the standards and guidelines...are not ready in time, we should not rule out postponing some parts of the AI Act,” Executive Vice President Henna Virkkunen said in Luxembourg during a meeting with EU digital ministers on Friday, according to Politico.
The first-ever AI-governing proposed act was drafted in April 2021 and received mixed reviews from industry experts. Although the move to establish rules over the use of AI has been welcomed by EU Member States' governing bodies, many warned of regulatory overreach and uncertain interpretations.
The first restrictions came into force in February and will continue to gradually roll out over the next year and a half.
Providers of sophisticated AI models will have to comply with the new rules from August 2nd, but the most significant code of practice is not yet ready.
Since the drafting of the original act in April 2021, the European Commission has been criticized for a lack of certainty in the application of the stringent compliance requirements and a potentially high barrier to entry for small companies.
EU Member States and partners are also starting to raise concerns about the act due to changes in administrations. Particularly, some worry that excessive regulations can kill innovation in AI, while others are anxious about the speed at which key provisions are being negotiated.
“There are concerns about how the EU AI Act might prevent exponential innovation in AI in Europe, compared to the strides being made by US and Chinese companies and governments,” said Nitish Mittal, partner at a research firm, Everest Group.
The US Mission to the EU has also offered its feedback on the draft, suggesting streamlining it and deleting certain provisions. The negotiations between the US and EU on the draft and general AI regulations are likely to continue as both parties try to avoid wider effects on trade.
The news comes as the UK government signalled it would not force tech companies to disclose how they train their AI models.
Your email address will not be published. Required fields are markedmarked