Ed Newton-Rex

The UK government's 'Sovereign AI Unit' has partnerships with 4 companies.

All are North American.
All are being sued for alleged copyright infringement.

It is beyond astonishing that this is who the government saw fit to partner with.

www.gov.uk/government/c...

I don't think so, no

As a judge recently said, it is likely already illegal in many cases, and unlikely to qualify as ‘fair use’. But this would put an end to the debate.

“It's time for Congress to give the American worker their day in court to protect their personal data and creative works.”

2/2

Senators Hawley & Blumenthal are introducing a bill that would definitively make it illegal to train AI on copyrighted works without permission.

🧵 1/2

www.axios.com/pro/tech-pol...

“Let’s just grab all the books we can for AI training, what could possibly go wro-”

chatgptiseatingtheworld.com/2025/07/17/a...

Creator organisations should canvas the views of their members on this issue, rather than lobbying for things that make the organisation more indispensable but that harm creators.

4/4

IMO these orgs are working in support of their own interests, not creators'. And they are doing this without the knowledge of the people they represent.

I implore you to contact your representatives in industry and ask them for specifics about what they are lobbying for.

3/4

I hear there are some orgs who are meant to represent creators who are lobbying for compulsory licensing, the proceeds of which they themselves would distribute.

But compulsory licensing would force you to let AI companies use your work to compete with you.

2/4

🚨 If you're any sort of creative in the UK, get in touch with the organisations that represent you and ask what they are suggesting the government do about AI & copyright.

There are some orgs who are meant to represent creators who are lobbying for their own interests.

🧵 1/4

They will be outcompeted by the AI models the government plans to legalise, and most of them will miss the chance to opt out.

Independent creators are the most important people the government should be hearing from.

3/3

The government's proposal - a broad copyright exception for AI training, which creators can opt out of - is bad for all rights holders, but it will hit independent creators the hardest.

2/3

Here is the list of creators' representatives the UK government included in their first working group on AI & copyright today.

It is mostly big media organisations - there is virtually no representation of smaller and independent creatives.

🧵 1/3

I'm not sure this is the right solution - but it's great to see the Commission coming out so strongly in favour of respecting creators' rights, and making it clear that tech companies' dreams of unpaid use of copyrighted works cannot be permitted.

Full Report: www.europarl.europa.eu/RegData/etud...

- AI systems don't learn like humans, from a technical or legal perspective

It suggests a solution of a statutory exception for AI training with an unwaivable right of equitable remuneration for authors and rights holders...

3/n

- Current EU law does *not* permit generative AI training on copyrighted work without a license. Some claim Article 4 of the CDSM does, but this view is mistaken.
- Requiring creators to opt-out or have their work trained on for free would be unfair and contravene international law

2/n

Incredibly important report on AI training & copyright released today by the European Commission. It gets lots right, slamming the idea that creators should be required to opt-out of having their works used for free by AI companies.

It says:

- Generative AI must be based on opt-in consent

🧵 1/n

If the government were serious about regulating AI companies, it would not be making those companies' products an ever more integral part of its own systems.

The government should be clamping down on copyright infringement by AI companies, rather than legitimising it in this way.

/end

Google trains its AI models on British creators' work without permission. Why is the government working with Google to accelerate its own adoption of these models?

3/n

A month ago, the UK government refused to make AI companies like Google reveal the copyrighted work they train their products on. Today, it announced a partnership that will see Google train government employees to use those very products.

2/n

More evidence today that the UK government values American AI companies more highly than the rights of British creators.

www.gov.uk/government/n...

🧵 1/n

JSTOR launched an AI reading assistant that seems to be powered by OpenAI’s models, which are trained on writers’ work without permission.

@jstor.bsky.social can you confirm?

I believe that can instead refile

This is a much more thoughtful interpretation of copyright law than yesterday's decision, and I suspect time will show it is the correct one.

/end

"... Which means that the companies, to avoid liability for copyright infringement, will generally need to pay copyright holders for the right to use their materials."

8/n

Judge Chhabria could not be more clear. "The upshot is that in many circumstances it will be illegal to copy copyright-protected works to train generative AI models without permission..."

7/n

... and accused him of "blowing off the most important factor in the fair use analysis" - the market effect on the work that is copied.

Tech lobbyists will frame the headline as "Meta wins on fair use", to try to convince people things are going tech companies' way. They are not.

6/n

He called out Judge Alsup's ruling yesterday in the Anthropic case, which went against authors on fair use, as being based on an "inapt analogy" (likening AI training to human learning)...

5/n

Generative AI has "the ability to severely harm the market for the works being copied, and thus severely undermine the incentive for human beings to create."

He went even further...

4/n

To the question of whether unlicensed training is illegal, he said "in most cases the answer will be yes".

He said generative AI can flood the market, undermining the market for the originals that are copied, disincentivizing creation.

3/n

Authors sued Meta for training on their books; Meta claimed fair use. Judge Chhabria actually ruled it was fair use. *But* he was clear: he only ruled this because he felt the authors argued the case badly.

He went further...

2/n