Ireland's Taoiseach (Prime Minister) had to respond to my piece in The Guardian yesterday morning.
His answer is incorrect: when Ireland fined Meta €1.2billion that was because it was ordered to.
In fact, the Irish DPC had proposed a fine of €0!
See para 38 of EDPB Decision 1/2023.
There was a time when Europe looked like the world’s answer to the worst excesses of big tech. Ireland has sabotaged that dream for a massive payoff.
Ireland collected almost 5x corporate tax per person than France or Germany (2022 is the latest figure I saw).
Tomorrow, Ireland takes over as EU Council Presidency for six months, when the EU is renegotiating tech and tax rules. In The Guardian this morning I argue it must recuse itself from these discussions because it is so captured by big tech. https://www.theguardian.com/commentisfree/2026/jun/30/ireland-big-tech-lapdog-eu-presidency-digital-sovereignty
I am very happy to announce that we have appointed Louise Hooper as Enforce's new AI expert. Louise was a barrister at top London set Garden Court Chambers for more than two decades, and has joined us full time.
A year ago today I said it would not be necessary to contemplate child social media ban if we had faith in our regulators or tech.
Strong tools exist but go unused:
-AI Act Art 1(b)
-GDPR Art 8 and 9
-AVMSD Art 28 (b)(1)
-DSA (interim measures, crisis response mechanism)
https://www.iccl.ie/digital-data/iccl-enforce-calls-for-digital-enforcement-at-irish-bar/
The fantastic Caoilfhionn Gallagher KC is interviewing former President of Ireland Mary McAleese on stage at ICCL’s 50th anniversary conference right now.
We start at 11 o’clock, court 3, four courts
3 day preliminary hearing in ICCL class action v Microsoft tomorrow. In Ireland’s first ever class action, this hearing asks: who bears the burden of proof when big tech uses your data in bizarre & hidden ways? Is it you or they who is required to be transparent about it? https://enforce.ie/digital-data/preliminary-hearing-on-burden-of-proof/
3 day preliminary hearing in ICCL class action v Microsoft starts tomorrow. In Ireland’s first ever class action, this hearing asks: who bears the burden of proof when big tech uses your data in bizarre & hidden ways? Is it you or they who is required to be transparent about it?
Posted without comment
I’m making a text adventure game.
You wake up on a bench in 2026.
You are Paul Baran.
Your last memory is 1970.
The game is called “A little caution”.
I told Reuters last week that Meta's new employee-surveillance-to-train-AI-to-replace-employees (apparently called the "Model Capability Initiative") is a sign of what is to come every worker, everywhere.
We want to hear from Meta employees.
I have been appointed Adjunct Full Professor by University College Dublin!
"U.S. forces deployed to war zones have been targeted using commercially available location data"
Just like I, @johnnyryan and others warned.
US Senator Wyden says it's time to "start treating the adtech industry as a national security threat". Agreed.
https://www.reuters.com/business/media-telecom/pentagon-says-us-military-personnel-are-reportedly-being-targeted-using-location-2026-05-28/
Reminder: Adtech is dangerous. New story, same problem.
https://www.reuters.com/business/media-telecom/pentagon-says-us-military-personnel-are-reportedly-being-targeted-using-location-2026-05-28/
9 years since I blew the whistle. 8 years since formal complaint to Irish DPC (which supervises Google and others across all of Europe).
Senator Wyden's letter today --> https://www.documentcloud.org/documents/28168364-ron-wydens-may-28-2026-letter-to-the-department-of-defense/..
I spoke to @alejandrotauber publisher of @euobserver about how Europe's "Ireland problem" could lead to tech disaster for our children, industry, and politics https://euobserver.com/218368/dublin-cant-trust-itself-to-do-the-right-thing-why-the-next-irish-eu-presidency-should-recuse-itself-from-tech-talks/
China's robust regulatory safeguards have not stifled its AI innovation.
In recent months Chinese AI has advanced faster than US competitors, and the U.S. NIST reports that they are already more cost efficient.
Europe should reflect on China’s performance and deregulation = innovation assumptions.
The chart shows the latest composite AI (LLM) performance data. China’s top LLMs (Qwen 3.6 Max and Kimi 2.6) are now within 5% of the top US model (ChatGPT 5.5). In contrast, Europe’s best performing model is the Mistral Medium 3, which lags China by 12% and the US by 17%.
AI race: US leads. How far behind are China and Europe?
Our plot of 2 years of AI Epoch Capability Index scores (a composite of LLM test scores)
Note: EU firms acquired by US firms counted as US.
Data source https://epoch.ai/data/benchmark_data.zip (20 May 2026)
