Transatlantic AI cooperation is essential for our climate, health and security

The following is an excerpt from my speech to the Congressional AI Caucus on April 27 in Washington DC.

Thank you, Congressman McNerney, for having me here today.

It’s a great honour to be speaking to you in Congress. I’ve been to the US many times, and worked with and for many US firms. Your drive to innovate and always be better is what has made you a global tech powerhouse. There is so much that Europe can learn from your entrepreneurial mindset.

AI has the potential to solve some of the world’s biggest challenges, starting from climate change. 40% of global CO2 emissions come from buildings. Digital twins, smart sensors, and AI technologies play a crucial role in monitoring and reducing our energy consumption across our traditional sectors.

AI is increasingly becoming key to strengthen our cyber security too, by detecting risks faster and more precisely, before even any attacks happen.

And in digital health applications,?AI is helping diagnose early stage cancer with an accuracy that no human can achieve.

Helping AI innovators grow on both sides of the Atlantic

The EU and US share common goals and a common history. We also share an economic relationship worth $6 trillion in commercial sales each year, supporting 16 million jobs.

But we share so much more. Our democratic values bind us together, as does our respect for individual freedoms.

But even between two like-minded democracies, regulatory barriers mean that smaller companies find it harder to expand. That means that citizens are not getting access to the latest and best technologies. And we’re missing out on growth.

AI innovators are already facing difficulty with divergent regulation in different markets. Oncompass Medicine, from Hungary, use AI to support physicians to deliver tailored cancer treatment to patients.

They are now expanding to the US, hopefully delivering this life-saving treatment to American cancer sufferers and contributing to the US economy at the same time. This is the kind of win-win digital trade and investment we like to see.

Because of the complexity of medical device regulations, they already have to go through two sets of procedures. This is a huge burden on a small company. AI in healthcare is so new, and there is a lot of uncertainty. Joint standards and a common understanding of concepts, risks and applications would be extremely beneficial.

As initiatives to regulate AI proliferate, let’s not add to that burden. We have a huge opportunity here – we are at the beginning of the process to align our approaches and minimize red tape. This is how we maximise innovation?and make sure the best ideas get the chance to deliver.

It is not just about improved AI and machine learning applications:?alignment will mean we can set the standard for the responsible use of AI. I do want us to agree on what we don’t want it to be used for. Social scoring, for example, or to discriminate against people based on their ethnicity or sexuality.

Regulating AI: where do we stand?

Globally,?at least 60 countries have adopted AI strategies, sometimes with regulation planned. We are all facing the same challenges?but are seeing a patchwork of different approaches pop up globally.

On regulation, what the EU does matters.

In scale and impact, the AI Act could be as important as GDPR. Although well intentioned, it risks being burdensome for companies to put into practice. But there is a possible silver lining: the EU wants to use technical standards to underpin regulation this time around.

The best way to use this opportunity is the Trade and Technology Council, which is meeting next month in Paris for the second time.

The TTC is the best chance we’ve got to align ourselves on AI and other issues. At the first meeting, the EU and US agreed on a risk-based approach to AI regulation. Let’s push further: for instance, let’s agree on common principles for assessing risk levels, so that the EU and the US do not diverge too much.

We must be prepared to present a united digital front

The war in Ukraine has made cooperation on AI and other emerging technologies essential. Warfare is increasingly digital, and AI is already being used on the battlefield. To win these hybrid wars, we need to stay one step ahead of our adversaries.

One aspect is digital skills. If we don’t have the skills, then we have a security risk for the future of democratic societies.

Aligning on standards also means that our technologies and protections are interoperable, and that when the next crisis comes, we are prepared to present a united digital front. If we don’t set the standards then someone else will.

I thank you for your attention.