In today’s column, I examine the recently proclaimed EU AI Continent Action Plan, announced in Brussels on April 9, 2025, and formally setting the stage for the EU to become a so-called AI Continent. Questions abound. Does the EU have what it takes to establish itself as an AI continental powerhouse? If the EU proceeds, will this spur other AI continental endeavors and trigger an all-out continent-by-continent rivalry?
Let’s talk about it.
This analysis of an innovative AI breakthrough is part of my ongoing Forbes column coverage on the latest in AI, including identifying and explaining various impactful AI complexities (see the link here).
EU Makes Bold AI Proclamation
In case you haven’t heard or seen the EU’s proclamation about becoming an AI Continent, I’d like to start by highlighting some key portions of their declaration (source: “Commission sets course for Europe’s AI leadership with an ambitious AI Continent Action Plan”, April 9, 2025, European Commission, Official Press Release). Several associated official documents go into detail about the AI Continent Action Plan, and you might want to read those if this is a topic of keen interest to you.
First, this is the overarching idea of what the EU intends to accomplish:
- “To become a global leader in artificial intelligence (AI) is the objective of the AI Continent Action Plan launched today. As set out by President von der Leyen at the AI Action Summit in February 2025 in Paris, this ambitious initiative is set to transform Europe’s strong traditional industries and its exceptional talent pool into powerful engines of AI innovation and acceleration.” (ibid).
You can readily compare the expressed sentiment to the same proclamations made by various national AI initiatives. Many individual nations have come forward and insisted that they will be a magnet for AI innovation, see my coverage at the link here.
Those nations though are embarking on their own pathway to do so. In this instance, the EU has opted to lift the table stakes and make this a continent-wide effort. It is an ambitious and perhaps gutsy move.
The Five Major Steps
One of the first thoughts you might have is what precisely the EU will do to become a said-to-be AI Continent.
You are right to be inquisitive.
There have been headline-grabbing declarations about AI on a nation-state level that often were vacuous. Other than claiming they wanted to be a haven for AI development, not much meat was to be found. Nothing burgers.
Some of those eye-catching proclamations have pretty much faded over time and little has ever been overtly carried out to achieve the pledged AI mission, see my analysis at the link here.
Anyway, the EU AI Continent Action Plan declaration outlined these five major steps:
- (1) Building a large-scale AI data and computing infrastructure.
- (2) Increasing access to large and high-quality data.
- (3) Developing algorithms and fostering AI adoption in strategic EU sectors.
- (4) Strengthening AI skills and talents.
- (5) Regulatory simplification.
Those steps certainly seem enticing.
Unpacking The Five Steps
In the first step, we already know that advancements in AI tend to require massive computing resources, thus building large-scale AI infrastructure makes abundant sense. Build it, and they will come (so goes the famous line).
Regarding the second step, generative AI and large language models (LLMs) are greatly driven by having access to suitable data, ergo if the EU could make high-quality data available readily and cheaply or freely, it would be a boon to LLMs.
On the matter of the third step, they want to foster wider use of AI. That’s fine and fits aptly to an AI Continent approach. But, according to the details in the proclamation, only 13.5% of companies in the EU have so far adopted AI. I’m a bit skeptical of that statistic. I believe it understates the on-the-ground reality. I can see why they would use such a stat since it bolsters the point of spreading AI usage wider across the EU.
Part of the issue is what counts as having adopted AI, such as whether you must build something in AI to count as an AI adopter or merely make use of a package that contains AI capabilities. Another aspect of counting is that oftentimes small businesses are the preponderance of the number of businesses all told, and they generally are slower to adopt new tech. This then undercounts or underweights the AI adoption by medium and large-sized firms.
The gist though of the third step is still of an uplifting nature, namely that fostering AI adoption throughout the EU and especially in strategic EU sectors would be a worthy element of becoming an AI Continent.
The fourth step of strengthening AI skills and talents throughout the EU would indubitably be a wise activity. You see, there is an ongoing worldwide battle for AI talent. Top AI talent tends to go where the opportunities and big bucks are best to be had.
Some nations worry they are facing a brain drain in that their pool of AI talent is being drawn to other quarters. According to the EU proclamation, they are aiming to further streamline legal migration pathways for highly skilled non-EU AI workers, and “attract the best European AI researchers and experts back to Europe.”
The Thorny Fifth Step
I saved the fifth step so that it could have its own moment in the sun.
Here’s the deal.
The fifth step says that regulatory simplification will be sought by the EU. Well, that portion of the proclamation has spurred quite an onslaught of chuckles and guffaws. The deal is this. The EU painstakingly proposed and passed the AI Act, which I covered step-by-step throughout my column coverage, see the link here.
The AI Act has a lot of controversy associated with it.
Some vehemently exhort that the AI Act is excessive regulatory oversight. Thus, having the EU now say that they are aiming for regulatory simplification is an outrageous and insulting claim from that perspective. It is as though this is spurious talk from both sides of the mouth. On the one hand, the considered “onerous” mandates of the AI Act are already in place, yet somehow regulatory simplification is on the horizon.
Hogwash, say skeptics.
How will the EU undertake the vaunted regulatory simplification?
Here’s what the EU proclamation stated:
- “The AI Act raises citizens’ trust in technology and provides investors and entrepreneurs with the legal certainty they need to scale up and deploy AI throughout Europe. The Commission will also launch the AI Act Service Desk, to help businesses comply with the AI Act. It will serve as the central point of contact and hub for information and guidance on the AI Act.” (ibid).
A cynical unpacking suggests that the regulatory simplification is made up of smoke and mirrors.
First, the claim is that the AI Act has already solidified the rules of the game, which then makes life easier for those who want to make and invest in AI in the EU. A contrarian would say that those rules reduce incentives for making and investing in AI in the EU due to their harsh and unyielding demands. In general, having a known playing field can be a handy means of encouraging AI activity, but if the playing field is stacked against the AI makers and AI investors, it does the opposite of calming things down. It discourages AI investments and tends to push them to more hospitable locales.
Second, the idea of an AI Act Service Desk is a nicety but seems not a likely means of moving the dial. Yes, providing assistance on what the AI Act consists of and how to comply is undoubtedly going to be helpful. The problem is that if you believe that the AI Act is a rat’s nest of bureaucracy and legal endangerment, no amount of assistance is going to solve the core issue. At the core, you’ve still got an anchor around your neck.
To be fair, not everyone perceives the AI Act in such draconian terms. Many believe the AI Act is a hallmark of what ought to be done to govern AI. They worry that other countries are letting the horse out of the barn. The rush toward unfettered innovation is going to shove unproven and unsafe AI down our collective throats and we will regret that we didn’t tighten the legal screws sooner.
Time will tell.
EU Seeks Public Input
Speaking of time, the timeline of the EU AI Continent Action Plan is stated as contingent on a series of initial public consultations.
A snippet of those is excerpted here:
- “With this Action Plan, the Commission opens today two public consultations, running until 4 June 2025, to further shape these AI Continent Action Plan initiatives.”
- “A public consultation inviting all interested parties to share their views on the Cloud and AI Development Act.”
- “A public consultation on Apply AI to identify stakeholder priorities, challenges to the uptake of AI, and the relevance of proposed solutions and policy approaches—including additional measures to ensure the smooth and simple application of the AI Act.”
- “A third public consultation on Data Union Strategy will be launched in May.”
- “In parallel, the Commission will organize dialogues with industry representatives and the public sector to help shape the Apply AI Strategy.”
Seeking public input is a welcome approach.
The question will be how the input is actively used to refine the Action Plan. Will the public input be fully leveraged or might it be summarily discarded? How will the public input be weighed in comparison to the input from the Commission and consulted experts in AI?
And so on.
An optimist would say that this is an aboveboard method that will ensure that the EU AI Continent ambitions are sensibly vetted and laid out for all to see. A vociferous doubter would contend that this is going to drag on for a long time, probably get snagged in paperwork and that you would be unwise to hold your breath about anytime soon witnessing a cohesive, practical, and down-to-earth EU AI Continent emerge.
The Big Picture
Why would the EU want to be an AI Continent?
Plenty of significant benefits would arise across the board.
First, predictions in general are that the adoption of AI is going to create tremendous economic growth for any nation willing to do so. If the EU can ensure that AI is widely and deeply adopted in the EU, the base assumption is that the EU would benefit accordingly from immense economic growth. The nations that are within the EU all become boats that rise with the rising tide.
Boom, drop the mic.
Another basis for wanting to gain widespread EU AI adoption is that doing so would presumably reduce dependence on AI from other parts of the globe. The EU would seemingly have sufficient AI capabilities within the confines of the EU. No need to rely on other non-EU nations for their AI. AI-related autonomy for the EU would be a notable prize. Sometimes this is referred to as digital sovereignty or AI sovereignty, see my analysis at the link here.
The other side of the coin is that not only would the EU be construed as AI-strong internally, but it would be in a grand position to sell or rent its AI capabilities to other entities beyond the EU. Think of it this way. The catchphrase “AI made in Europe” could become an amazing branding angle. Other countries and businesses outside of the EU might clamor for AI that has been built in the EU.
AI becomes a top-notch export from the EU and is heralded as being the best AI that money can buy.
Sour Views Of The Big Picture
That rosy perspective of a decidedly successful EU AI Continent being achieved must be balanced against the potential downsides and risks involved.
Perhaps the EU would lose focus and drop the ball on the AI Continent Action Plan.
The logic is as follows. This is a matter that is going to require dogged determination and persistence to bring to fruition. It is going to take many years to suitably devise and roll out the Action Plan. Energies might get depleted, and attention might shift to other priorities.
It all fizzles out.
There’s more to consider too.
Will the EU nations embrace the AI Continent scheme or prefer to go it alone on their respective national AI interests?
Internal fragmentation would almost certainly be the death knell for the AI continental ambitions of the EU. Too many cooks in the kitchen might produce lousy results. Lack of coordination and cooperation could lead to a scattergun outcome.
Wait for a second, you might be thinking, why can’t the EU nations be AI-national and simultaneously be AI-continental? Some would say that this is spreading things too thin. The result might be that the national AI interests are half-baked, and the AI Continent turns out to be half-baked as well.
Finally, one of the most vocal criticisms is that the EU would need to cave in on the AI Act and other pending regulatory proposals if they truly want to see the EU become an AI Continent. The only sensible pathway to being an AI Continent would be to promise that AI innovation is the highest priority. Given the EU’s regulatory efforts on AI so far, almost no one would see a future in which AI innovation overpowers AI regulations.
In short, the necessary ingredients of all-in AI innovation won’t be enmeshed in the AI Continent keystones, which, regrettably, ultimately the restraints and limits on AI will undermine the upbeat and rosy aspirations accordingly.
The Bigger Picture
Currently, you could reasonably contend that we are living in a multi-polar AI world.
Nation-states tend to be keeping their AI efforts close to their chests. Each nation wants to be the first to arrive at the revered pinnacle of AI such as artificial general intelligence (AGI) and artificial superintelligence (ASI), see my discussion at the link here. This makes clearcut sense. Imagine the geo-political power that a nation could wield if it had AGI or ASI, while other nations are still steeped in conventional AI.
Such a nation would leapfrog to the top of the world’s nations.
Now then, if a nation opted to share and openly bond with other nations during the AI journey toward AGI and ASI, they would inevitably need to likewise share in the pot of gold at the end of the rainbow. Why do so? If you believe your nation can muster the right stuff and attain AGI or ASI on its own, splitting the final prize just doesn’t comport.
Aha, the key here is whether nations of themselves have the wherewithal to reach the grand prize.
The EU AI Continent Action Plan might get some non-EU nations to rethink their go-it-alone mindset. Maybe the EU is onto a clever insight that deserves due attention. Individual nations might get left in the dust unless they are willing to partner with other like-minded nations.
A future scenario might consist of other AI Continents being formulated in response to the EU efforts.
Here are four substantive AI Continents that might arise:
- EU AI Continent: As per the EU AI Continent Action Plan.
- Silicon Americas AI Continent: Possibly the U.S., Canada, Mexico, and select Central and South American countries.
- Sinosphere AI Continent: China and other select countries.
- Pan-Indo AI Continent: India, Japan, and select ASEAN countries.
There are many ways to slice and dice which nations would seek to form an AI Continent with other nations. Since geography might not be the only factor involved, some refer to such conglomerations as AI Blocs rather than AI Continents.
AI Planet
A final twist for you to mull over is that some believe we need to be an AI Planet.
Say what?
Rather than each nation vying to attain AGI or ASI, and rather than nations forming AI Continents or AI Blocs, we need to come together as humans and realize that we are going to be an AI Planet.
AI will be the most crucial element for the future of humanity. Furthermore, we need to make sure that AI doesn’t rise to a level that it opts to wipe us out or enslave us, the dreaded existential risk of AI.
Therefore, the entire planet and all nations must work collaboratively on AI. We will all fall or rise once AGI or ASI is attained. Better to get our act together now, working side-by-side, as humans, and ensure that AI comes out as a thing of good versus something of dire evil.
Can all the nations of earth undertake a kumbaya oath and as one devise and shape where AI is going?
It’s a pretty tall order, that’s for sure.
