EU AI Act Monthly Digest
EU AI Act News: June 2026 Digest
By Derrick D. Jackson
~7 min read
This digest covers 7 key EU AI Act developments from June 2026. Each item is sourced and verified. For interactive compliance tools, visit the EU AI Act Compliance Hub.
Enforcement and Compliance
20260603
Industry Study Claims €600B at Risk in EU TDM Copyright Debate, What the Research Actually Represents
CCIA Europe, an industry association representing major technology companies, launched a study at the European AI Roundtable on Copyright on approximately June 3 projecting that restricting the EU’s text and data mining copyright exception could cost the EU economy up to €600 billion annually. The figures come from an industry-commissioned study and have not been independently verified; they’re designed to shape the European Commission’s cost-benefit analysis, and understanding their provenance is as important as knowing the numbers. (CCIA Europe)
20260602
Connecticut Signs PA 26-15: AI Companion Apps Must Detect Self-Harm and Restrict Minors by January 2027
Governor Ned Lamont signed Connecticut Public Act 26-15 on June 2, 2026, finalizing the state’s AI responsibility framework and making Connecticut the second state this week to impose specific legal requirements on AI companion applications. Two compliance deadlines now apply, October 1, 2026 for high-risk AI provisions and January 1, 2027 for companion AI restrictions. (Connecticut Governor’s Office / Connecticut General Assembly)
GPAI and Foundation Models
20260603
The EU AI Act Copyright Fault Line: TDM Rules, GPAI Compliance, and What AI Developers in Europe Face
Three separate pressure points converged on European AI developers this week: an industry lobby launched a €600 billion economic projection to push back against EU TDM copyright restrictions, the EU AI Office reportedly clarified GPAI compliance obligations in a new FAQ with eight weeks until the August 2 deadline, and the broader copyright litigation pattern in the registry confirms that rights holders aren’t waiting for legislation to arrive. Developers building or deploying AI in the European market aren’t navigating one policy question, they’re navigating three overlapping tracks that interact in ways the regulatory text doesn’t fully resolve. Understanding the fault line between them is more useful than monitoring any one track in isolation. (CCIA Europe)
20260602
EU AI Act Enforcement Bodies Now Staffed: Scientific Panel and Advisory Forum Formally Appointed
The European Commission officially appointed members of the EU AI Act’s Scientific Panel and Advisory Forum on June 1, 2026, giving the European AI Office the independent expert infrastructure needed to act on GPAI model compliance. For organizations still treating EU AI Act enforcement as a future problem, June 1 changed the calculus. (European Commission, Digital Strategy)
20260601
EU AI Act Digital Omnibus: A Compliance Obligation Map by Organization Type
The EU AI Act’s Digital Omnibus amendment package doesn’t change the law uniformly, it restructures obligations differently depending on whether your organization provides general-purpose AI models, deploys high-risk AI systems, or builds applications that generate user-facing content. Compliance programs built on the original text aren’t wrong; they’re incomplete in different ways depending on which track you’re on. This piece maps what changed, for whom, and what each organization type still needs to resolve before August 2. (Latham & Watkins / EU Commission / Inside Global Tech)
20260601
EU AI Act Digital Omnibus: Three Changes That Split Compliance Obligations by Organization Type
The EU AI Act’s Digital Omnibus amendment package restructures compliance obligations differently depending on whether you’re a GPAI provider, a high-risk system deployer, or a limited-risk application developer. The deadline clock hasn’t stopped, but what you’re racing to complete by August 2 has changed. (Latham & Watkins / Inside Global Tech)
20260601
Multi-Framework AI Cybersecurity Obligations for Financial Institutions: DFS, NIST, and the EU AI Act
DFS-regulated financial institutions now face AI cybersecurity obligations from
at least three directions simultaneously: New York’s Part 500 advisory framework,
NIST’s CAISI analysis finding existing controls insufficient for agentic AI threat
profiles, and the EU AI Act’s GPAI-SR security requirements for frontier model
providers. These frameworks were developed independently, but they converge on the
same gap in most organizations’ security programs: controls designed before frontier
AI existed don’t address what frontier AI enables. This piece maps who owns what
obligation across the three frameworks and identifies where the gaps compound. (New York DFS (Frontier AI Cybersecurity Advisory, May 21 2026))
Compliance Resources
Explore our interactive tools and downloadable guides for EU AI Act compliance.
The EU AI Act Monthly Digest is published by Tech Jacks Solutions.
Sources verified as of June 03, 2026. Updates published monthly.