EU’s fragmented AI regulatory framework is hampering innovation: Meta and Spotify CEOs

0

The European Union’s “fragmented regulatory structure” is stifling artificial intelligence (AI) innovation, Meta CEO Mark Zuckerberg and Spotify CEO Danie Ek said in a joint statement published on Friday. The two tech leaders added that the EU’s AI regulatory framework is “riddled with inconsistent implementation,” hampering the ability of the region’s companies to take advantage of the AI wave. 

They noted: 

“Instead of clear rules that inform and guide how companies do business across the continent, our industry faces overlapping regulations and inconsistent guidance on how to comply with them.”

The two founders warned that without “urgent changes” the EU will be left behind in the AI race. 

The case for open-source AI

According to the two founders, open-source AI, where models and tools are publicly accessible under permissive licenses, offers a significant opportunity for European organizations. Open-source AI can democratize access to advanced technologies, preventing the concentration of power among a few major players and fostering a more competitive and innovative environment. 

They pointed out that much of the internet and leading tech companies already rely on open-source technologies, suggesting that the future of AI development will similarly benefit from open collaboration and transparency. 

They stated: 

“We believe the next generation of ideas and startups will be built with open-source AI, because it lets developers incorporate the latest innovations at low cost and gives institutions more control over their data.”

Meta, in particular, has embraced this model by open-sourcing several of its AI technologies, including its Llama large language models. These models have already been used by public institutions and researchers to advance medical research and preserve endangered languages. 

Similarly, Ek attributed Spotify’s success to its utilization of AI at an early stage.

The CEOs emphasized that with Europe boasting a significant number of open-source developers, the continent is well-positioned to leverage this approach to AI development.

Regulatory challenges

According to the CEOs, while regulations are necessary, pre-emptive regulation of nascent technologies like AI will do more harm than good. They noted: 

“Europe’s risk-averse, complex regulation could prevent it from capitalising on the big bets that can translate into big rewards.”

The founders further expounded on the regulatory challenges facing European companies by pointing out the uneven application of the EU’s General Data Protection Regulation (GDPR) law. According to them, while the regulation aimed to “harmonise the use and flow of data” across the region, regulators now cannot make up their minds on how to apply the law. 

For instance, Meta has been directed to delay training its models on publicly shared content from Facebook and Instagram, until regulators figure out how to apply GDPR. In the short term, these delays are creating uncertainty. 

In the case of Meta, the delay means that AI models will not reflect the “collective knowledge, culture and languages of Europe.” The delay also prevents Europeans from using the latest AI products. For example, Meta will not be able to release its Llama model in the EU. 

The statement noted: 

“The stark reality is that laws designed to increase European sovereignty and competitiveness are achieving the opposite.”

A call for change

The CEOs claim that the EU’s inconsistent AI regulatory framework has already caused a brain drain with most AI developers working outside the region. This trend will only accelerate and the region will miss out on a “once-in-a-generation” opportunity unless Europe actively changes its approach. 

“Europe needs a new approach with clearer policies and more consistent enforcement,” they noted. A simplified regulatory regime will not only accelerate the growth of open-source AI but also offer support to European developers, they stated.

The statement noted: 

“Though Spotify and Meta use AI in different ways, we agree that thoughtful, clear and consistent regulation can foster competition and innovation while also protecting people and giving them access to new technologies that empower them.”

Mentioned in this article

Credit: Source link

Leave A Reply

Your email address will not be published.