EU AI Act Expands Scope of Application
Claire Weston
The EU AI Act begins enforcing transparency rules and general-purpose AI model regulations on August 2 — chatbots must disclose they are AI, deepfakes must be labeled. Some high-risk rules are delayed to 2027–2028, but compliance obligations for general-purpose model providers tighten immediately.
What rules take effect on August 2?
The European Commission announced on July 31 that the AI Act will expand its enforcement scope from August 2, activating two sets of rules: transparency obligations and general-purpose AI (GPAI) model provider rules.
This means → the Act entered into force earlier, but most provisions were not yet enforceable. August 2 is when regulation gains teeth.
The European AI Office also gains enforcement authority over GPAI model providers from the same date.
What do the transparency rules require?
Chatbots and other interactive AI systems must inform users they are talking to AI, not a human.
AI-generated or AI-modified images, video, and audio — deepfakes — must be clearly labeled and carry machine-readable markers for automated detection.
In plain terms = if a hyper-realistic video crosses your feed, the platform must tag it as AI-made. No tag, no compliance.
The Commission says these rules aim to reduce deception and manipulation.
What new obligations face general-purpose AI providers?
Providers of advanced GPAI models that may pose systemic risk must meet additional requirements.
Those requirements cover four risk categories: cyberattacks, loss of model control, harmful manipulation, and violations of fundamental rights.
This means → companies like OpenAI and Google operating large models in the EU must demonstrate they have safeguards in place — or face enforcement action.
What is the full timeline of the Act?
The AI Act entered into force on August 1, 2024, classifying AI systems into four risk tiers: unacceptable, high, transparency, and minimal-or-no risk.
It follows a phased enforcement schedule; most provisions apply from August 2, 2026.
In plain terms = the Act does not switch on all at once — it rolls out in waves, tackling the most urgent obligations first.
Which high-risk rules have been delayed?
High-risk AI rules for recruitment, education, law enforcement, and immigration are postponed to December 2, 2027.
High-risk AI embedded in regulated products such as medical devices and machinery is postponed to August 2, 2028.
This reflects a deliberate trade-off: the EU tightens GPAI provider oversight immediately, but grants one to two extra years of buffer for sectors requiring deep industry integration.
Content is for reference only, not financial advice.