the Court of Appeal massively rejects the High Court’s finding in RTM v Bonne Terre (Sky Bet) that marketing/cookie consent might be invalid as the gambling firm is aware this person is a vulnerable gambler close to the regulatory exclusion threshold. caselaw.nationalarchives.gov.uk/ewca/civ/202...

Interesting read on how business models and imaginaries of future engineering tasks are competing, and which AI firms are placing bets on different directions to act as their new platform ‘moat’ thenewstack.io/ai-agent-har...

Many voters who use big, old platforms may rarely see age verification practices and not be aware of invasiveness, as such platforms (Apple already does) use age of account as a proxy and never check more. Younger users get no such benefit. good for data minimisation, but keeps people in a bubble.

Some injunctive success against ICE tracking app store takedowns pressured by the US DoJ. Law not very good at constraining implicit state power and threats though, esp where plaintiff isn’t the threatened party. Incentive to kiss the ring can be too great. www.theverge.com/policy/91461...

i’ve moved everything to rss and built a load of automatic parsers for non rss sites and ngos and think tanks and whatever and a way to rss any journal that issues dois and it’s so calm and peaceful

Great to see @eurosky.social launch access to PDS accounts for AT Protocol hosted entirely on EU servers today. We were pleased to be able to support the initiative with legal advice on EU data/platform law in its early stages. www.euronews.com/next/2026/04... eurosky.tech

"Defunct startups are being liquidated for their Slack archives, Jira tickets, and email threads—operational exhaust that AI labs now treat as premium training data."

If Anthropic's Mythos is 'too dangerous to release', then we have to ask (as cybersecurity firm Aisle did): why did smaller, open-source models also find the same vulnerabilities as Anthropic reported as revealed by their 'frontier' capabilities? aisle.com/blog/ai-cybe...

Client-side scanning comes to 3D printing, with proposals to have printers forced to embed algorithms to scan for weapons. Printing weapons is a bad thing, but so is ubiquitous surveillance from your own devices. A dangerous model for other areas. www.eff.org/deeplinks/20...

“Homework help” service Chegg fined $500K in federal court under Australia’s academic cheating law www.austlii.edu.au/cgi-bin/view... A couple of yrs back we argued these laws are so broad, some arguably make providing generic AI systems a criminal offence doi.org/10.1080/1360...

I wrote up my concerns about the CJEU's Russmedia ruling. This is the first of two posts, it's about the bad results of relying on the GDPR (and displacing laws like the DSA) to define platforms' liability for content. Post two will review arguments that the case is more limited. 1/

US now turning to grand juries as a potential way — circumventing immediate first amendment arguments — to force intermediaries like Reddit to de-anonymise users who are criticising government policy theintercept.com/2026/04/10/r...

Apple user being advised data is unrecoverable after iOS update removes the ability to enter certain Czech accented characters, which are in his password (a háček), from the lock screen keyboard. www.theregister.com/2026/04/12/i...

Apple's lack of iOS documentation on what happens within its operating system creates real issues when the threat moves to the hardest zone to defend against — a compromised device. Their Platform Security documentation should consider this threat model too. help.apple.com/pdf/security...

excellent, if depressing, takedown of the recent HEPI report on uni marketisation and its ‘solutions’ @jimdickinson.bsky.social wonkhe.com/wonk-corner/...

TIL: recent iOS updates are locking cops out of seized iPhones. For months now, forensic tools like Cellebrite haven’t been able to extract data from some of these devices One workaround I’ve heard: police will subpoena jail phone systems to get defendants PINs…then try those codes on the phones

I’ve posted online our 2025-6 20 week Internet Law and Policy syllabus on our LLB programme, run by Orla Lynskey, Tommaso Fia and me. We hope it’s useful for other tech law educators! michae.lv/updated-inte...

Full time, permanent law and politics lecturer post at @laws.ucl.ac.uk. Background either a social scientist with law training, or a lawyer with social science interest. Teaching: either/both of introductory EU/PIL, plus specialist teaching. Civil lawyers welcome! www.ucl.ac.uk/work-at-ucl/...

After 5 months, the European Commission has now released several of the industry consultation responses provided in relation to the AI Act's General Purpose Model provisions www.asktheeu.org/request/all_...

This dude is rotating inspirational quotes in his email signature. Intense total dipshit energy.

🚨CFP: The ADM+S Generative Authenticity project is hosting a small scholarly symposium on Authenticity, Generative AI and Synthetic Media on 6-7 May at QUT Kelvin Grove (and hybrid node). Short abstracts due 9 March. All info and the submission form: docs.google.com/document/d/1...

A Haaretz investigation has found that at least three Israeli cyber companies are currently operating in #CARINT; one has developed an "offensive" tool that can potentially tap into your 🚗car's microphones and cameras. www.haaretz.com/israel-news/...

Hardly needs to be said, but this rule would be gibberish and unenforceable if everyone didn't already know what topics and perspectives are going to be ruled "controversial"

sophia scam robot maker in the epstein files? say it ain’t so!

Ben Goertzel, popularizer of the term AGI and creator of the Sophia robot, congratulating Jeffrey Epstein the day after Epstein was released from prison. Goertzel is mentioned in the files nearly 800 times, including well after the Miami Herald investigation in 2018.

I'm continuing my research into global tech politics, including the politics of digital sovereignty, and the political economy of the digitalisation and statecraft in my role at ODI Global - please reach out if you're keen to collaborate or want to know more about my research!

I’ll be putting out much more of this work into the world in the weeks and months ahead, but for now, I want to thank my supervisor @michae.lv for his mentorship and support, and my examiners Dr. Gavin Sullivan and Dr. Cecilia Rikap for their excellent feedback and critique.

My thesis examines the mutual interactions between emerging platform and data-based technologies with forms of political economic authority, structures and subjectivities, through a case study of India’s ‘Digital Public Infrastructure’, aka the 'India Stack'.

Last week, I successfully defended my PhD Thesis, ‘Platform Politics and Data Publics: Sovereignty, Statecraft and Citizenship in the India Stack’, at the Faculty of Laws, University College London.

Financial barriers to environmental litigation are against the law, because protecting the environment is an important public interest. The government now wants to make it super expensive (>€100,000) for ordinary people to go to court to protect the environment Please read and take action 👇

Over on Twitter/X, Bill Ackman (of Harvard/Claudine Gay/plagiarism/tennis debacle fame) has written another 3,000+ word post. He believes that Francesca Gino's RAs faked her data. And, plot twist, he's been funding Gino's legal battle against Harvard since 2024. x.com/i/status/200...

Happy New Year www.whatdotheyknow.com/request/inco...

As you’re watching a Disney film at Christmas, raise a glass to Disney’s lawyers who, in Florida’s 2021 social media law, had a blanket exemption placed in to except any theme park owner from the definition of ‘social media platform’ www.flsenate.gov/Session/Bill...

Our new paper demonstrates that removing images of children from training datasets of text-to-image models fails to prevent the misuse of these models to generate child sexual abuse material (AI CSAM). A thread 🧵

We spent long time studying the question "does training data filtering help prevent generative models from generating unwanted content?" TL;DR: Nope. In some cases it slightly increases difficulty but not enough to prevent even average users from finding ways to create what they want.

Europe’s Digital Markets Act aims to open ecosystems: alternatives to app stores, fair access to OS features, real interoperability. But for many free and open source developers, those promises remain fragile without robust enforcement, write Lucas Lasota, Dario Presutti and Jithendra Palepu.

Tech whistleblowers say coming forward has unexpectedly derailed their lives... They became isolated among their colleagues, suffered severe professional damage, or were pushed out of the industry altogether... www.washingtonpost.com/technology/2... by @naominix.bsky.social

Well, this is an utter disaster for science. This gutting of peer review is very likely driven by Project 2025/OMB/Vought, very likely is about funding junk science, and very likely is coming for #NIH and biomedical research next. 🧪

NSF under threat from itself.

VPNs starting to enter the regulatory frame: now in Denmark www.theregister.com/2025/12/15/d...

Not sure the Online Safety Acts — either UK or AUS flavours — are ready for kids setting up and running social media services themselves as regulated entities. This already happens: expect more. I ran online forums *as a child* that would now need child access assessments under the UK OSA!

EC has formally opened an investigation into Google for using publishers’ content without meaningful consent or compensation for its AI Overviews feature. Big win for civil society orgs like @foxglovelegal.bsky.social who’ve been asking for this. ec.europa.eu/commission/p...

One thing I found odd was the reference early in the judgment to the content being manifestly unlawful. But if it was, the identity checks that the controller has to carry out wouldn’t be necessary. FWIW, it comes from para 17 of the referring court’s request. curia.europa.eu/juris/showPd...

it's a little known fact but the AI Act bans academic articles containing the words "in the Era of" — sorry if that's you, I don't make the rules

Russmedia case is out — CJEU says it's fine to put on data protection obligations onto an advertising platform to spot sensitive data and verify who is in it. But would this create an obligation to monitor that would break intermediary liability shielding?

Useful if terrifying read on the Chinese brand of smart watches for kids, Little Genius, which is forming an entire, insular — and seemingly pretty socially unhealthy — platformised and corporatised world. substack.com/@ivyyang/p-1...

Never come across 'WindowsCentral' as a media outlet but this piece is powerful and dense and absolutely spot on about where we are with AI and business models right now. Putting bigger outlets to shame. www.windowscentral.com/artificial-i...

"They're trying to change our habits, because all of the projections rely on people becoming truly dependent on the technology. Whether or not it's actually a good thing for society isn't considered to be a factor."