Dragoş TUDORACHE
Hoping that companies will self-regulate is not enough to safeguard our citizens in our societies. And we are now hearing more and more of the digital giants out there fearing their own products. […] What took me by surprise and by storm is generative AI based on powerful foundation models. […] ensuring that the models potentially used by millions of smaller companies and start-ups, as well as by hundreds of millions of individual users, that they are safe.
Artificial Intelligence Act (debate)official record ↗
Read the full statement ↓Hide the full statement ↑
Mr President, dear Commissioners, dear colleagues, as we move into this debate, I want to address the one question that I got the most over the past four years, working with many of you in the AIDA Committee and then working on this file. Is this the right time? Is this the right time for Europe to regulate AI?
And my question is resolutely ‘yes’. It is the right time because of the profound impact the AI has. And I am convinced that all of you today will be bringing these arguments to the fore. It is the right time also because hoping that companies will self-regulate is not enough to safeguard our citizens in our societies. And we are now hearing more and more of the digital giants out there fearing their own products. It is also the right time because we’ve done preparations and worked seriously, responsibly on this file.
And I want to start by thanking the Commission for having the foresight already some years ago to start preparing these rules and put them forward. I want to thank the leadership of all political groups and the leadership of this House for having the foresight to establish a Special Committee on Artificial Intelligence, because we are now better prepared as lawmakers and I think we have ever been, to actually tackle these rules.
And now let me move on to the rules themselves. What will we vote tomorrow? I will walk you through some of the key elements in many thanks to my co-rapporteur Brando with whom I worked very well for already pointing some out, I will try not to repeat them.
We have worked hard to provide a definition of AI that is not only agreed by all political groups in this House, but also aligned with that of the OECD and other partners such as the US. This gives the democratic world a starting point for developing a common terminology when discussing AI.
We are providing thorough texts developed in parliaments, start-ups and SMEs in Europe with a strong network of sandboxes so they can innovate, they can grow and develop technology in accordance with European values and in respect of fundamental rights, increasing the trust in their products as well as their competitiveness.
We have worked to put in place a coherent governance structure that makes sense. It is meant to ensure that we consolidate the digital single market, that Member States have the proper tools and support in place for the enforcement of this regulation, that Europe has a future-proof approach to new developments in AI and that we do not repeat some of the mistakes we have done with GDPR.
Ten or even five years down the line it is this governance structure that will give Europe the ability to deal with the rapid evolution of AI and to reap the most benefits from it. And we have worked first and foremost to ensure our citizens’ rights and freedoms are not just respected but protected and strengthened. We don’t want mass surveillance. We don’t want social scoring. We don’t want predictive policing in the European Union, full stop.
That’s what China does, not us. We want all public authorities, whether they distribute benefits or whether they enforce the law to be accountable, transparent and focused on our rights with strict democratic oversight.
Last but not least, I am proud that we had the foresight to devise a preliminary set of rules for AI. What took me by surprise and by storm is generative AI based on powerful foundation models. We worked on rules that allow technology to flourish and develop while ensuring that the models potentially used by millions of smaller companies and start-ups, as well as by hundreds of millions of individual users, that they are safe. We introduced obligations of diligence for content generated by these models and of transparency regarding copyrighted material used in their training. And very importantly, we found overwhelming consensus among all political groups on these rules. And that, I think, gives a very powerful statement,
Colleagues, I say, taking full responsibility for my words, we are making history. We are writing history. We are shaping the global development of AI in the right direction with humankind and human interests at the very centre.
Thank you very much. And I really want to give a special thanks to all the shadow rapporteurs for how we have come together around this text that we will vote on tomorrow.
Text as published by the European Parliament, in the Parliament's own English rendering of the original language.