The EU AI Act has entered a new phase of enforcement, bringing key phases of the bloc’s comprehensive AI regulations into practical application and placing new obligations on both AI developers and organizations deploying the technology.
From August 2, 2026, key provisions covering general-purpose AI models became enforceable, expanding regulatory oversight beyond how AI applications are used to include how some of the underlying models powering them are developed and managed.
But developers are not the only ones facing new obligations. The latest measures also introduce new transparency requirements across the AI ecosystem, with businesses required to provide clearer disclosure around AI usage, including when customers are interacting with AI systems or when content has been generated or manipulated using the technology.
What The New AI Act Rules Mean For AI Developers And Businesses
The AI Act’s latest requirements place new obligations on providers of general-purpose AI models, including companies developing large language models that can be adapted for a wide range of use cases. The European Commission’s AI Office will oversee enforcement for general-purpose AI models.
Under the rules, AI developers must provide greater transparency around how models are built, including information about training processes and the data used to develop them. Providers of the most advanced “frontier” models will face additional obligations to identify and mitigate wider risks linked to their systems.
However, the impact of the legislation extends beyond AI developers. Organizations using AI tools in customer service, communications, content creation, and internal operations will also face new responsibilities, particularly around transparency.
Businesses must make clear when users are interacting with AI rather than a human, while AI-generated or manipulated content, including deepfakes, may need to be labeled. Certain content will also require machine-readable markers to help identify whether it has been created or altered using AI.
Companies that fail to meet the requirements could face penalties of up to 3% of annual global turnover, increasing pressure on organizations using AI services across European markets.
How Enterprises Can Prepare For AI Act Compliance
For enterprises, the introduction of the transparency rules means AI governance must become part of everyday operations.
The first step for businesses should be understanding where AI is already being used. Organizations adopting AI-powered customer service platforms, workplace assistants, and content generation tools will need visibility into where these systems are deployed across the business. For systems already deployed, companies have until December 2 to make the required changes. New AI applications must meet the stricter transparency requirements immediately from this coming month.
Customer-facing AI deployments will require particular attention. Organizations introducing AI-powered chatbots, virtual assistants, or automated communications will need to ensure users understand when they are engaging with technology rather than a human employee.
Companies should also assess whether their AI suppliers provide the controls needed to support compliance. While businesses might be able to apply disclosures informing customers that they are interacting with AI, they may not have the capability to add machine-readable markers if they are relying on a third-party provider.
Although these new regulations can seem like a burden, Alexandros Koronakis, Senior Director, Head of Government, Policy & Regulatory Affairs, EMEA, Twilio, believes the new rules represent an opportunity for forward-thinking companies:
“The AI Act’s transparency obligations coming into force is much more than a regulatory issue to manage. It is an opportunity to redefine the customer experience, operationalizing a vision where AI functions in service of trust and eliminates friction to drive a superior overall CX,” Koronakis said.
“In a context where digital trust is becoming a strategic asset, the leaders who anticipate do not merely aim to be compliant. They set the standard.”
How The AI Act Could Shape Global AI Governance
The enforcement of the AI Act represents a major milestone in global AI regulation, with the EU attempting to create a framework that could influence how companies develop and deploy the technology worldwide.
Because the legislation applies to organizations offering AI services within the European market, its impact will extend beyond European borders, affecting international AI developers and global enterprises serving EU customers.
However, some in the industry have warned that transparency requirements are only one part of the wider AI governance challenge. Nik Kairinos, CEO and Co-founder of RAIDS AI, said businesses should not wait for future regulatory deadlines before strengthening their AI safeguards:
“Transparency is a start, but continuous oversight that monitors AI behavior in production, after deployment, is what will determine whether AI can be trusted at scale.”
While developers will face greater scrutiny over the models they create, enterprises will need to focus on how those models are used and whether customers and employees are given sufficient transparency.
The challenge for businesses, many of which are still moving AI from pilot programs into wider operational use, will be keeping track of regulatory requirements as both the technology and their own use cases continue to evolve.
