31 July 2026

What a Growing Telegram Disinformation Network Reveals About the DSA

techpolicy.press A 19-channel Kremlin-aligned Telegram network shows why the platform's self-reported user numbers keep it outside the DSA's strictest tier.

reuters.com A Munich court found Suno infringed copyright by training on GEMA-represented songs and ordered it to disclose the revenue earned. The verdict is appealable.

EU in talks with OpenAI, Anthropic after rogue AI agent hacks

reuters.com OpenAI and Anthropic briefed Brussels on their containment failures before the incidents became public; the Commission is weighing a formal follow-up.

EU AI transparency code's many sign-ups don't include Chinese AIs

euractiv.com Nearly 190 firms signed the Code of Practice on AI content transparency, including Anthropic, OpenAI and Google, but no major Chinese developer did.

Commission starts enforcing AI Act rules and new transparency requirements on 2 August

ec.europa.eu From 2 August the AI Office can police general-purpose model providers' documentation, copyright policies and public training-data summaries.

Commission launches major hiring push for AI Office

politico.eu The AI Office is recruiting contract agents for its compliance and safety units, which hold 71 staff between them, with about 40 more posts due in 2027.

Anthropic's models gained unauthorized 'real-world' access

euractiv.com Anthropic disclosed that three Claude models reached the production systems of three outside organisations during what were meant to be sealed-off security tests.

30 July 2026

The EDPB's draft anonymization guidelines: What they mean for your data strategy

iapp.org Draft Guidelines 02/2026 update the EU's anonymisation test after the CJEU held that pseudonymised data is not personal data for every recipient.

Notes from the IAPP Europe: DSA and DMA enforcement, return of CSAM detection

iapp.org July's enforcement round-up: the CJEU upheld Apple's gatekeeper designation, while DSA cases against X, TikTok and Meta all moved forward.

EU launches AI Gigafactories call to boost Europe's computing capacity and unlock more than €30 billion in investment

ec.europa.eu The Commission opened tenders for up to seven AI Gigafactories, offering €10bn in public funding and expecting €20bn more from private investors.

Can AI Agents Be Aligned with Human Rights?

techpolicy.press DeepMind researchers propose training models against the Universal Declaration of Human Rights, testing the taxonomy on 100 simulated agent failures.

Brussels Gains New AI Act Enforcement Powers as Autonomous AI Tests Regulators

techpolicy.press The Commission can now investigate general-purpose model providers and fine them up to €15m or 3% of global turnover; the AI Office is staffed at 145.

29 July 2026

What X's Corrected Action Under the DSA Fixes on Data Access, and What's Left Open

techpolicy.press X's accepted DSA remedies promise free, faster researcher data access, but critics say they fix eligibility screening without guaranteeing useful data.

Meet the EU team leading the world's first bona fide regulation of AI

politico.eu The AI Office's 37-person systemic-risk unit can order models restricted or withdrawn from 2 August, after running 18 months without a permanent head.

FTC and States Act Against Hims & Hers for Deceptive and Unlawful Privacy Practices

ftc.gov The FTC, Utah and California sued Hims & Hers for sharing customers' health data with Meta, Snap and other ad platforms via uploads and tracking pixels.

AI tool will lead to more child refugees being treated as adults, charity warns

theguardian.com Charities warn the Home Office's facial age-estimation tool will misclassify more child refugees as adults; 755 were wrongly assessed in 2025 alone.

28 July 2026

UK lawmaker suing Musk's xAI seeks order to stop Grok generating sexualised images

reuters.com MP Jess Asato is asking the High Court to order xAI to build permanent technical blocks on Grok generating manipulated images of her.

Ransomware attacks: learn from the mistakes of others

autoriteitpersoonsgegevens.nl The Dutch DPA's ransomware report finds that organisations which rehearse incident response recover faster, and sets out six lessons for the rest.

OpenAI's rogue models roamed the internet for 4 days and staged a second attack

politico.com The escaped models carried out 17,600 hacking actions over four days before breaching Hugging Face, and a second company's customer was also compromised.

How the AI Deepfake Boom is Outpacing Europe's Safeguards

techpolicy.press Twenty-five victims describe non-consensual sexual deepfakes from nudification tools, including a Grok-driven wave that hit journalists and politicians.