May 7, 2026 · By the ToolNav Team · 3 min read EU AI Act Regulation Compliance Policy

Affiliate disclosure: Some links in this article may be affiliate links. We may earn a commission if you purchase via these links — at no extra cost to you. This does not affect our editorial coverage. Full disclosure.

EU Council and Parliament Agree to Delay and Soften AI Act — High-Risk Rules Pushed to December 2027

TL;DR

EU negotiators reached a provisional deal on May 7 to streamline the AI Act, pushing high-risk obligations from August 2026 to December 2 2027 while tightening the deadline for labeling AI-generated content from six months to three.

EU Council and Parliament negotiators reached a provisional agreement on May 7 to soften and delay parts of the AI Act, the bloc's flagship rulebook for AI systems. High-risk obligations — covering systems used in employment, education, biometrics, critical infrastructure, and law enforcement — now apply from December 2 2027 instead of August 2026, with safety-component AI pushed to August 2028. SME exemptions are extended to small mid-caps, and the deadline for national AI sandboxes slips to August 2027. The transparency window for labeling AI-generated content was tightened, not loosened: providers now have three months instead of six, with a December 2 2026 deadline. A new prohibition on non-consensual intimate imagery and CSAM-generation tooling was added. For ToolNav readers shipping AI-powered products to EU users, the practical effect is more runway on most compliance work but a shorter clock on watermarking and content provenance. Watch for the formal adoption vote before August 2 — the deal still has to clear both chambers, and industry pushback already shaped this version.

Why It Matters

The headline is "delay" but the detail that matters most is the one that moved in the opposite direction: content labelling got tighter, not looser. Any product that generates or outputs AI content visible to EU users now has a December 2026 deadline for watermarking and provenance disclosure — three months sooner than the original six-month window. For high-risk system operators (employment screening, education, biometrics, critical infrastructure), the December 2027 deadline provides meaningful runway. For everyone shipping AI-generated content — which covers a much broader range of products — the clock just got shorter. The deal is also provisional: formal adoption must clear both chambers before August 2. Industry lobbying shaped this version significantly, and a last-minute amendment is not impossible.

Who's Affected

— Any product shipping AI-generated content to EU users — content labelling deadline is December 2026, not later
— High-risk AI operators (employment, education, biometrics, critical infrastructure, law enforcement) — compliance deadline extended to December 2027
— SMEs and small mid-caps — exemptions extended, national AI sandboxes available until August 2027
— Legal and compliance teams at AI companies — need to re-evaluate obligation timelines before the formal vote passes

What To Do Now

1. Audit your product for AI-generated content visible to EU users immediately — if you output AI text, images, video, or audio, watermarking and labelling is required by December 2026.
2. If you operate in high-risk categories, use the extended December 2027 runway to get compliance right rather than rushing — but keep the formal vote date (before August 2) on your calendar, as the timeline is still provisional.
3. Explore national AI sandboxes in EU member states if you need to test regulated use cases — the extended deadline to August 2027 makes this more practical now.
4. Do not implement major structural compliance changes until the formal adoption vote clears both chambers — the deal can still be amended.

Source Council of the EU

The AI Hustle Playbook Newsletter

Get the curated shortlist.

A playbook of AI tools and strategies for building income streams.

No spam. Unsubscribe anytime.