AI Act and Digital Omnibus: How European Simplification Is Reshaping Obligations for American Companies

The European Union is modifying its regulatory framework while companies are already in operation. On May 7, 2026, the European Parliament and the Council of the EU reached a provisional agreement on the so-called Digital Omnibus — the legislative package with which Brussels intends to simplify the implementation of the AI Act and recalibrate certain key deadlines.

For US companies operating or intending to operate in the European market, this development is not a technical detail: it is an element that directly affects strategic planning.

What Is the Digital Omnibus Changing?

The Digital Omnibus does not represent a step back from the AI Act — it is, more accurately, a targeted regulatory adjustment designed to make the regulation more actionable, in light of certain objective delays and operational complexity that emerged during the initial implementation phase.

On the one hand, the technical standards developed by European standardization bodies are not yet fully available; on the other, several Member States have not yet completed the designation of national supervisory authorities or the establishment of genuinely operational structures.

Added to this is the fact that the actual timelines for adapting high-risk systems are proving longer than initially estimated.

On a practical level, the most significant result for companies is the prospect of a postponement of obligations for high-risk systems, with a shift of certain deadlines from August 2026 to late 2027 or 2028, depending on the category of system.

Alongside this, extensions are also foreseen for certain AI-generated content watermarking obligations (now due by 2 December 2026) and an expansion of simplification measures for small mid-cap companies (SMCs).

The Trap of Waiting for the Digital Omnibus

The main risk introduced by the Digital Omnibus is, paradoxically, that of inaction. A full compliance process for high-risk AI systems under the AI Act requires on average many months of work — covering system mapping, risk analysis, control design, documentation preparation, and engagement with conformity assessment bodies.

Postponing the start of these projects until final deadlines are confirmed often means ensuring a last-minute scramble.

The notified bodies responsible for conformity assessments are already facing a significant volume of requests, with the concrete risk of bottlenecks as key dates approach.

For US companies, the postponement does not reduce the complexity of the journey — it merely shifts its concentration to a later point on the timeline.

Those who have already initiated internal governance processes, system classification, and technical documentation preparation have not “jumped the gun”; on the contrary, they are in a position of advantage, because they will be able to use the postponement window to consolidate, test, and optimize what has already been set up.

The Digital Omnibus Should Not Be Treated as a Regulatory Amnesty

There is one aspect of the Digital Omnibus that deserves particular attention from companies looking at the European market.

The simplification pursued by the European legislator does not consist in reducing substantive protections or attenuating essential requirements, but in making obligations more proportionate, procedures clearer, and governance more operational.

In other words, the aim is to bring the regulation closer to concrete operational realities, without modifying its fundamental structure.

This is the same principle that should guide a professional approach to corporate compliance: not simplifying the regulation itself, but simplifying its reading, decomposition, and translation into processes, roles, controls, and documents.

Angela Lo Giudice, lawyer at Polimeni.Legal and professional active in the fields of privacy, GDPR, and digital law, frames the issue as follows:

“The Digital Omnibus confirms that even the European legislator recognizes the distance between the written norm and its practical implementation. For US companies, this is not a signal to slow down, but to structure more effectively. Regulatory simplification only works if there is already an internal governance framework to lean on. A company that arrives unprepared for a simplified rule remains unprepared all the same. If a company uses the deadline extension to put its processes in order — inventorying systems, defining responsibilities, documenting decision flows and controls — then simplification becomes a force multiplier. If, on the contrary, the postponement is read as an opportunity to delay further, the result is concentrating more risk into less time, precisely when the regulatory framework becomes fully operational.”

This perspective clarifies an essential point: simplification is an opportunity only for those who have already begun to build solid foundations.

The “Brussels Effect” and Global Implications

The combination of the AI Act and the Digital Omnibus takes on additional significance when read in light of the so-called “Brussels Effect” — that capacity of the European Union to influence, through its standards, regulation and market practices at the global level.

The AI Act is not only affecting the European perimeter: other jurisdictions are developing regulatory frameworks inspired by or otherwise attentive to the logic of risk classification, transparency, and operator accountability.

For a US company, conforming to the AI Act therefore means not only gaining access to the European market, but also preparing for standards that tend to resonate across various legal contexts — reducing the need to chase disparate regulations country by country.

The compliance framework built for Europe can become an exportable governance foundation, capable of engaging with future national or state-level AI rules.

From Regulation to Simplification: The Real Competitive Advantage

The European intervention should not be interpreted as a mere bureaucratic adjustment — it is better understood as a political and market signal.

The European legislator is showing it wants to accompany companies on a realistic adaptation path, without abandoning the underlying objective: making artificial intelligence systems safer, more transparent, and more governable.

For US companies, this solution could translate into a concrete competitive advantage.

A company that, in advance, builds a clear AI governance framework — with defined roles, structured processes, updated registers, and coherent technical documentation — is able to respond more promptly to requests from European authorities and partners, to navigate due diligence processes, vendor selection, and internal audits with less friction, and to present its compliance as a differentiating factor in commercial negotiations — especially in regulated sectors or those with high reputational sensitivity.

The Digital Omnibus ultimately introduces a new question for companies already engaging with the AI Act: is simplification an opportunity or a risk? The answer depends on how the journey has been approached so far.

For those who have treated compliance as a structured process — system inventory, qualification of the role in the supply chain, risk classification, technical documentation, definition of responsibilities and decision flows — the deadline extension is the opportunity to consolidate, refine, and make AI governance scalable.

For those who have instead adopted a wait-and-see approach pending clarifications, guidelines, or simplifications, the risk is that even clearer rules arrive too late relative to operational needs.

The true competitive advantage does not lie in formal compliance with deadlines, but in the capacity to turn AI governance into a stable organizational competency, independent of fluctuations in the regulatory calendar.

For US companies looking to Europe, the AI Act and Digital Omnibus are not merely a terrain of regulatory risk — they are a proving ground for demonstrating maturity and the capacity to integrate innovation within a well-defined framework of accountability.