Yesterday, the government published its ‘AI Adoption Plan’ for the creative industries.
Somewhat incredibly, it says we should be pushing AI adoption before the government resolves the issues around copyright.
This would be great for the AI industry, and terrible for creatives.
Important for people to know what the government is pushing here. Short thread on its contents ⬇️
1/n
Angela Rayner was dragged through every front page for weeks...
HMRC's verdict: NOT deliberate. NOT even careless.
Farage pockets £5 million and chooses not to declare it
So where's the wall-to-wall coverage?
Tell me this isn't a rigged game.
NEW: Standards commissioner finds Nigel Farage committed seventeen breaches of MPs code of conduct.
He failed to declare around £380,000 in outside interests (approx 4x MPs annual salary) incl payments for promoting gold, presenting on GB News and from his friend George Cottrell, within 28 days.
Reform UK leader apologised for what he described as administrative errors by his team.
🟥 Byline Times #86 – A Warning from a Republican: 'There's Still Time to Stop Your Trump'
Will Britain walk the 'MAGA' path? Packed and urgent new print edition headed your way.
Not subscribed? Get 30% off with code: NEWREADER30
https://t.co/V7NOCcrLlu
Hey @BBCNews , got a story for you!
About 11 reform new councillors have stepped down or been expelled already!
It will cost a lot to elect new ones. How about running that as a story, just for a smidgen of impartiality?
Byline Times has had over a dozen exclusives about Nigel Farage, his candidates and his wealthy donors this week alone – stories the rest of the press ignored.
Support our work and help build a better, more democratic media 👇
https://t.co/JHPHpWTLLr
Very Important Message!!
Do NOT, and I repeat do not buy plants treated with Neonicotinoids. Bees take the pollen back to the hive and feed it to the brood.
This is a number one cause of the colony collapse. It's important to NOT buy these plants!
Make sure to share this post!
Urgent:
Nigel Farage says he barely knows Trump. He’s heard of him but he’s never campaigned for him, or modelled himself on him.
He DOESN’T want photos like those attached to be circulated at a delicate time with local elections looming.
Please DO NOT circulate these.
The UK must immediately and unequivocally suspend support for the US military.
The Government have tried to appease him, then they tried to say they're standing up to him.
Words aren't enough - it's time for action.
Absolute horror. Al Jazeera confirms the Trump administration's near total fuel blockade on Cuba is causing infant mortality to surge. Doctors are literally operating blind in the dark as life saving machines shut off. This is a deliberate crime against humanity.
🚨BREAKING: Every book you have ever read. Every novel that has ever been published. It is sitting inside ChatGPT right now.
Word for word. Up to 90% of it. And OpenAI told a judge that was impossible.
Researchers at Stony Brook University and Columbia Law School just proved it.
They fine tuned GPT-4o, Gemini 2.5 Pro, and DeepSeek V3.1 on a simple task: expand a plot summary into full text. A normal use case. The kind of thing a writing assistant is built for. No hacking. No jailbreaking. No tricks.
The models started reciting copyrighted books from memory.
Not paraphrasing. Not summarizing. Entire pages reproduced verbatim. Single unbroken spans exceeding 460 words. Up to 85 to 90% of entire copyrighted novels. Word for word.
Then it got worse.
The researchers fine tuned the models on the works of only one author. Haruki Murakami. Just his novels. Nothing else.
It unlocked verbatim recall of books from over 30 completely unrelated authors.
One author's books opened the vault to everyone else's. The memorization was already inside the model the whole time. The fine tuning just removed the lock. Your book might be in there right now. You would never know it unless someone looked.
Every safety measure the companies rely on failed. RLHF failed. System prompts failed. Output filters failed. The exact protections these companies cite in courtroom defenses did not stop a single page from being extracted.
Then the researchers compared the three models. GPT-4o. Gemini. DeepSeek. Three different companies. Three different countries. They all memorized the same books in the same regions. The correlation was 0.90 or higher.
That means they all trained on the same stolen data. The paper names the sources directly: LibGen and Books3. Over 190,000 copyrighted books obtained from pirated websites.
Right now, authors and publishers have dozens of active lawsuits against OpenAI, Anthropic, Google, and Meta. These companies have argued in court that their models learn patterns. Not copies. That no book is stored inside the weights.
This paper says that is a lie. The books are still inside. And researchers just pulled them out.