Meta and Spotify have defended open source artificial intelligence (AI) as a way to boost the development of this technology in the European Union and prevent it from being concentrated in the hands of a few actors, in a plea that criticises recent European laws that limit them.
Meta and Spotify founders and CEOs Mark Zuckerberg and Daniel Ek have jointly discussed what they believe Europe needs to take advantage of the AI push without hindering innovation and competition.
In their editorial, first published in The Economist and later on the Meta blog, they highlight the role of open source technologies in the development of artificial intelligence, as they “allow developers to incorporate the latest innovations at low cost and offer institutions greater control over their data.”
“It is the best option to leverage AI to drive progress and create economic opportunities and security for all,” they add. In the current context, they point out, there is already a gap between those who have and those who do not have access to this technology, and for this reason they believe that open source can help ensure that “power is not concentrated in a few large players and, as happened with the Internet before it, creates a level playing field.”
However, they also understand that, although necessary, the regulation that has been approved in recent years in the EU, “complex and risk-averse”, can prevent technology companies from “taking advantage of big bets that can translate into big rewards.”
In this regard, they have charged against the General Data Protection Regulation, because under it Meta has had to delay “the training of its models with the content shared publicly by adults on Facebook and Instagram, not because any law has been violated, but because the regulators have not agreed on how to proceed.”
The complaint follows Meta’s plans to use information shared by users on social media platforms Facebook and Instagram – both owned by Meta – to train its AI models, plans it had to halt in mid-June after receiving a request from the Irish Data Protection Commission, in cooperation with its EU data protection authorities.
Before changing its plans, Meta provided an objection form for users to express their opposition to the use of their data for AI training. The data it intended to collect included information related to activity on the platforms, friends and followers, data about the applications, browsers and devices used or those coming from providers and third parties.
In the editorial, the two leaders insisted that this stance has led to the multimodal Llama model capable of understanding images not being available to users in the European Union.
They therefore say that Europe “needs a new approach with clearer policies and more consistent application.” “Although Spotify and Meta use AI in different ways, we agree that thoughtful, clear and consistent regulation can foster competition and innovation, while protecting people and giving them access to new technologies that empower them,” they conclude.