How a US AI Company Wins Customers in Europe, Beyond EU AI Act Compliance
Does the EU AI Act apply to a US company?
Yes, in most cases. The Act applies to companies that place AI systems on the EU market or whose AI output is used in the EU, wherever the company is based.
It sorts AI into levels of risk.
- Prohibited uses, such as certain forms of social scoring and manipulation. These bans already apply.
- General-purpose AI models. Obligations for providers of these models already apply.
- High-risk systems, such as AI used in hiring, credit decisions, education or critical infrastructure. These carry the heaviest duties. Under changes agreed in 2026, the dates were pushed back to December 2027 for standalone high-risk systems and August 2028 for AI built into regulated products.
- Limited-risk systems, such as chatbots and generated content, which carry transparency duties.
- Minimal risk, which covers most business software, with no specific duties.
The timeline has already changed once. Check the current dates with a lawyer before you plan around them. This article is not legal advice.
Why compliance is not a go-to-market plan
Almost everything written for US AI companies about Europe is about compliance. That answers "are we allowed to sell?" It does not answer "why would a European customer choose us?"
I have seen this from both sides. At Google I led abuse prevention for more than a billion accounts using machine learning. At compensIT I designed a deep learning model that made decisions about people's income. In both cases the technology was the easy part to explain. Trust was the hard part.
What European buyers ask about AI
Expect these questions early, often from legal, security and works council representatives, not only from the business buyer.
- Where does our data go? Is it used to train your models? Is it stored in the EU?
- How does the system reach its output? Can a person understand and challenge a decision?
- Who is responsible when it is wrong? What is in the contract?
- Which risk category are you in? And what have you done about it?
- Is there human oversight? Especially for decisions that affect employees or customers.
If your sales team cannot answer these clearly, the deal goes to a competitor who can, even one with a weaker product.
Turn the answers into sales material
Prepare a short trust pack before you launch.
- A plain-language description of what the AI does and does not do.
- Your risk classification under the AI Act and the reasoning behind it.
- A data sheet covering what you collect, where it is stored, and whether it trains your models.
- The human oversight and appeal process.
- Security certifications and your data processing agreement.
Hand it over in the first meeting. It shortens the sales cycle and signals that you took Europe seriously.
Choosing where to start
For AI products, the first market depends on the buyer more than the country. The UK is outside the EU and has its own lighter approach to AI rules, which makes it a quick start. Germany and France have large enterprise buyers who ask the hardest questions. Win there and your trust pack will work anywhere in Europe.
Common questions
Does the EU AI Act apply to US startups?
Yes, if the startup sells an AI system in the EU or its output is used in the EU. Company size and location do not exempt it, though there are lighter requirements for smaller companies in some areas.
Is my AI product high-risk under the EU AI Act?
It depends on the use, not the technology. AI used for hiring, credit scoring, education, essential services or law enforcement is typically high-risk. A coding assistant or a marketing tool usually is not. Get a legal view on your specific use case.
Should a US AI company launch in the UK before the EU?
Often yes. The UK is English-speaking and outside the AI Act, so first customers come faster. Plan for the EU from the start, because requirements there will shape your product and contracts.