Posted by harel 11 hours ago
"The government has cited a significant data protection breach as the reason for its decision - an issue it clearly has a duty to take seriously."
https://www.nuj.org.uk/resource/nuj-responds-to-order-for-th...
ETA: They didn't ship data off to e.g. ChatGPT. They hired a subcontractor to build them a secure AI service.
Details in this comment:
https://news.ycombinator.com/item?id=47035141
leading to this:
https://endaleahy.substack.com/p/what-the-minister-said
The government is behaving disgracefully.
They don't have a budget for that. And besides, it might be an externalized service, because self hosting is so 90s.
Traced what? Innuendo is not a substitute for information.
you _really_ shouldn't be allowed to train on information without having a copyright license explicitly allowing it
"publicly available" isn't the same as "anyone can do whatever they want with it", just anyone can read it/use it for research
Quoting from an urgent question in the House of Lords a few days ago:
> HMCTS was working to expand and improve the service by creating a new data licence agreement with Courtsdesk and others to expand access to justice. It was in the course of making that arrangement with Courtsdesk that data protection issues came to light. What has arisen is that this private company has been sharing private, personal and legally sensitive information with a third-party AI company, including potentially the addresses and dates of birth of defendants and victims. That is a direct breach of our agreement with Courtsdesk, which the Conservatives negotiated.
https://hansard.parliament.uk/Commons/2026-02-10/debates/037...
Ministry of Justice in UK has always struck me as very savvy, from my work in the UK civic tech scene. They're quite self-aware, and I assume this is more pro-social than it might seem.
If you don't "know about them from another source" you can't effectively find/access the information and you might not even know that there is something you really should know about.
The service bridged the gap by providing a feed about what is potentially relevant for you depending on your filters etc.
This mean with the change:
- a lot of research/statistics are impossible to do/create
- journalists are prone to only learning about potentially very relevant cases happening, when it's they are already over and no one was there to cover it