YOUR BRAND · YOUR CUSTOMERS · YOUR AI BUSINESS
Sell AI employees under your own brand.
Legal & Regulatory Frameworks Weekly AI News
July 20 - July 28, 2026Weekly signal
The EU moved from guidance toward immediate compliance signals for agentic and interactive AI this week (20–28 July 2026): the European Commission published practical Guidelines and a Code of Practice for the AI Act's Article 50 (transparency) on 20 July, and the Digital Omnibus amendment to the AI Act was published in the Official Journal on 24 July (entering into force 27 July). Those steps clarify when and how systems that behave like agents must disclose themselves and when AI‑generated content must be marked—rules that apply to many agentic use cases and to providers/deployers operating in or targeting the EU market.
What changed
-
Commission guidance: On 20 July the European Commission published Guidelines interpreting Article 50 (transparency obligations) and an accompanying Code of Practice focused on marking and machine‑readable signalling for AI‑generated content. The Guidelines take an explicitly agentic reading (interactive systems and systems that may interact with natural persons must disclose that they are AI; deployers must flag certain AI‑generated content and emotion/biometric categorisation). The Commission states these transparency obligations will apply from 2 August 2026.
-
Digital Omnibus entered into force: The targeted amendment package (Regulation (EU) 2026/1744, the “Digital Omnibus on AI”) was published in the Official Journal on 24 July and entered into force on 27 July. It delays some high‑risk Annex III obligations (new implementation deadlines, e.g., for certain conformity assessment requirements) while leaving the Article 50 transparency timeline in place—so marking and disclosure now have immediate legal force for many interactive/agentic deployments.
-
UK parliamentary attention and cross‑regulatory work: UK Parliamentary debates (House of Lords, 23 July) foregrounded agentic AI risks—loss of control, sandbox escapes, and cross‑sector legal gaps—pushing the need for coordinated policy across consumer, competition, data protection and safety regulators. The DRCF and UK government workstreams (calls for evidence on data regulation and consumer AI risk) continue to signal that the UK will address agentic AI via sectoral coordination rather than a single new statute.
What to do with it
- If you provide or deploy AI agents that can interact with people in or target the EU: implement Article 50 disclosures (user‑facing disclosure + machine‑readable marks) before 2 August 2026 and document your decisions in a compliance file. Treat the Commission Guidelines and the Code of Practice as the practical interpretive baseline.
- Reassess product timelines: the Digital Omnibus grants breathing room for some high‑risk procedural requirements, but it does not postpone transparency obligations—separate your roadmap for (a) immediate disclosure/marking and (b) longer‑term conformity work for high‑risk agentic systems.
- Update governance and audit trails: make runtime logging, identity signals (agent identity / operator contact), and human‑in‑the‑loop controls first‑class features of agents; regulators and parliaments now emphasise traceability and the ability to stop or audit agent actions.
- Engage with cross‑regulatory processes: respond to DRCF / GOV.UK calls and EU consultation channels; signatory codes (GPAI/Code of Practice) and participation in standards/ITU workstreams will reduce enforcement risk and provide interoperability benefits.
Sources: see numbered list in the sources array.
YOUR BRAND · YOUR CUSTOMERS · YOUR AI BUSINESS
Sell AI employees under your own brand.
Give businesses AI teams that own ongoing roles. Set your pricing, delegate routine coordination and build a recurring service.