The Council of the EU gave final approval on June 29 to the AI Act simplification package, the so-called Digital AI Omnibus, closing a fast-moving legislative sprint that began with the European Parliament’s endorsement on June 16 and is expected to land in the Official Journal in July, weeks before the original August 2 compliance cliff.

The headline concession: high-risk AI obligations move from August 2, 2026 to December 2, 2027, with regulated-product carve-outs (think lifts, toys) pushed further to August 2, 2028. Brussels is quietly conceding what DLA Piper, Latham & Watkins, and other observers have flagged for months, that harmonized standards likely wouldn’t arrive before the deadline they were supposed to govern.

Article 50 transparency obligations still bite on August 2, 2026. Generative systems already on the market must meet watermarking rules by December 2, 2026, the same day the prohibition on “nudifier” apps and CSAM-generating systems takes effect. Maximum fines remain €35 million or 7% of global turnover.

The real structural shift is downstream. The SME compliance-lite framework now stretches up to 750 employees and €150M in revenue, folding in reduced fines, sandbox access, and standardized templates. Tooling vendors, LemonLime among them, are already positioning for that expanded mid-market envelope.

Across the Atlantic, Reps. Jay Obernolte (R-CA) and Lori Trahan (D-MA) released the discussion draft of the Great American AI Act on June 4, proposing three years of state-law preemption. Goodwin notes the timing is pointed: Colorado has replaced its AI Act with the broader ADMT Act, and Illinois SB 315, adding independent third-party auditing, sits on the governor’s desk.

Brussels bought time. Washington is trying to buy silence.

Sources

Sources