A new report which supports AI looting is parking a think tank on publishing's lawn
A new UK report Rebooting Copyright argues that old copyright laws are dead in the AI era. It has many flaws, but the worst seems to be its assumption that creative lightning can be bottled.
What strange and sinister shadow is this now being cast upon our meadow of happily grazing publishers? Why, it's a copy of "Rebooting Copyright: How the UK Can Be a Global Leader in the Arts and AI" from the Tony Blair Institute for Global Change! It's a whopper at over 15,600-words, but do not fear, for what it has in size, it lacks in weight.
For those interested in endurance tests, the entire thing is here, and despite reading like many a proposed government policy document, we should remind ourselves that for now it is no more an official stance on the matter than the opinion of the Downing Street cat.
I should declare a personal bias here, although I would term it a learning experience. Many moons ago in my professional capacity as a reporter, I attended a "town hall" style political gathering for health workers, with the then-leader of the opposition, one Tony Blair MP, as the star attraction.
And it was there I witnessed the Blair Magic. Facing a hall full of hostile Glaswegian nurses, who'd heard all the promises before, the soon-to-be-Prime-Minister listened to them, engaged with them, smoothed their ruffled feathers, and somehow, somehow, managed to tell them he didn't really accept any of their points, but most of them left the venue carrying a feeling like he still did, really.
That's some magic.
Other than carrying the name of his grandiosely-titled institute, this new report into the UK's copyright laws and creative industries, its creative wealth, and the promised untapped wealth of tomorrow's AI utopia, does rather lack his political subtlety.
It does contain language such as "advance a multi-pillar transparency approach", so rather like a thesis with little to say, but produced to the most perfect standards of academic writing, it has the feel of weight, rather than the actuality. I would dare say it has the feel of something produced by AI, but that would be unkind to its authors.
I will say that, like many a proposed government policy document, much of what it says is perfectly reasonable and simply stating the emerging and likely realities at work - whether we like them or not.
But, also like many a proposed government policy document, key parts of it are not based in reality, and come across as a verbose exercise in sounding reasonable. Depressingly, key points are seemingly anchored in the weakest of grounds: the Fear Of Missing Out.
It has already drawn much reaction from industry figures, gathered together at rapid pace by Graham Lovelace here. And for an excellent and specific refutation of some other of the report's details, Ed Newton-Rex has done a great job here.
To cut a long story short, whatever language it uses and whatever examples of gold-flaked porridge tomorrow it promises, the bit that most obviously makes it feel like a rebadged tech company RFP is its strong backing of the "opt out" proposal which would give AI companies default free access to copyrighted work for AI training purposes, with the obligation on those who own or created the work to opt out of its being picked up by AI for training.
In terms of paradigm-shifting arguments, I find none in the document. In fact, in very many instances where the term opt-out is used, you could substitute "opt-in" and change little of what else is said in that section.
Too much AI is Average Intelligence, not Advanced Intelligence
One thing that stands out to me in the document is the conflation of machine-based systems with human creativity, and that pouring more into the bucket somehow guarantees a purer swill.
Maybe you have read Sam Altman's attempt to convince us that ChatGPT had successfully produced a piece of actual creative writing? Here's a refresher.
If that's your idea of truly creative writing, then I humbly suggest reading is not your thing, and that a career in rock mastication beckons. While it has something of the feel of the human, it lacks soul and direction, a simulacrum of creativity. That's because it is an aggregation.
There is already much research that - like the wisdom of crowds - the intelligence of LLMs in the wide open field of writing and interpreting the human condition does not get better just because you pour in more content. Especially if it regurgitated from itself, which we know is happening.
This is a sensitive topic to us at Glide, where over the last couple of years we have seen from many publishers and tech colleagues an honest desire for AI as Advanced Intelligence, and their frustration at the regression towards Average Intelligence that many of these bucket-of-everything models actually deliver in raw form.
It ain't easy being creative, and what they are most definitely not getting from AI is something creatively new.
Rebooting Copyright uses one example from fairly recent British artistic history to illustrate its central point that all "creativity", human or machine, is derivative in some way, and says: "... artists visit galleries, often with no entrance fee, to explore a variety of creative works. Was Tracey Emin expected to reimburse Louise Bourgeois for the transformative experience she had upon encountering her work at the Tate in 1995? The relationship between originality and imitation has always been ambivalent, from classical art to the present day."
You have to love the "often with no entrance fee" don't you? The direction of suggestion is clear. It really all reads like a robot landing on Earth and using physics to compute happiness, desire, or yearning.
To cut down Tracey Emin to one important artistic experience in framing your argument is to render everything else that found expression in her creativity as lesser. Of course, if we are to suggest equivalence between the creative totality of Tracey Emin and something that can produce Studio Ghibli lookalikes, then one fears the intellectual waters are shallow indeed.
We can only use the highest standards of creativity to set rules on the use of original content, or else we risk bringing such human endeavour down to the level of an AI-generated email response. Would Shakespeare have improved the originality the quality of his work if he had an AI quill? Would politicians using WhatsApp for policy conversations make better decisions if they had AI in the app? Well, turns out they now do.
Likewise, the Blair Institute report uses the example of the widespread adoption of photography and its subsequent influence on art, causing "a profound crisis in modern art" yet sparking "a series of transformative movements" as an illustration of its expectation of the progression of AI.
At present, and despite frequent and loud proclamations of promise such as this report, I do not see the evidence for this belief in the creative field. The internet itself, with its world-changing abilities in data, speed and communication, remains the transformative technology of our age, with smartphones coming hard in its wake.
Generative AI imagery for example is simply an aggregational filter, capable of some impressive results, but none that bring the fundamental changes to our lives that the internet itself has.
Finally, and as if to absolutely confirm the political instincts that sit at the heart of Rebooting Copyright, it proposes the establishment of the Centre for AI and the Creative Industries, CACI, in the UK.
To quote: "CACI would institutionalise the intersection of arts, industry, technology and policy. Led by international experts in creative AI, the centre would directly address current and future challenges facing the sector."
Sounds good doesn't it? They always do. They might as well also build a facility for bottling lightning next door.
Unlock the potential of your audience data with the audience management platform you've been waiting for
Bring authentication, entitlements, and preference management together into one actionable CDP.
Unify audience profiles across platforms and channels
Model users and groups
Build up and enrich profiles to power your monetisation strategies
See what Glide Nexa can do for you. Request your demo today.
Audience engagement Q&A with GPP!
GPP's barely-in-the-news duo Denis Haman and Rich Fairbairn sat down with State of Digital Publishing (SODP) to chat about first-party data, audience management, and Glide Nexa. With well over 50 years of media and publishing experience between them - yes, really - they had plenty to say in their typically reticent style.
Read
SEO rules? For moi?
There's a reason publishers don't really like machine translations without any human input: Google says it will penalise it. (It's one of the reasons our translation tools in Glide CMS work as they do.) So surely a ranking surge by fully-automated Reddit translations is an error soon to be rectified? Unlikely, given how cozy the pair are. Another asterisk in the SEO rulebook!
Read
Google News changes
News publishers: seen swings in your site's Google News visibility? Check for recent updates in your Publisher Center. AI is taking more of a role in how it shows off your work, and removed many config options you had to control how your work was displayed.
Read
March 2025 core update
SEO guru Barry Adams takes a look at the impact of Google's March 2025 core update rollout. ''Fairly uneventful'' for UK news sites.
Read
Ranking fluctuations
Wonder why new topics on your site sometimes rank well, then drop away? A Googler explains.
Read
SEO weighs in on GEO
GEO is an emerging field, but one that’s generating plenty of interest. There’s no accepted standard for GEO yet, though it will likely follow the same best practices as SEO: a well-structured site, clear content categorisation, content quality, and easy navigation. The Athletic’s Shelby Blackley breaks down why it matters, how publishers should approach it, and teases upcoming GEO tactics.
Read
LLMs.txt
Meanwhile, Australian technologist Jeremy Howard has proposed a new standard for letting your content be crawled and used by large language models - LLMs.txt. While the proposal has generated interest, it has its critics.
Read
Magical creations or a privacy nightmare?
The latest trend of AI-generated Ghibli-style artwork may be charming social media users, but it’s also setting off privacy alarm bells left and right. Experts warn of the risks - data breaches, misuse of personal information, and the potential exposure of sensitive data - all because these images will inevitably be fed back into AI training models. So, is a cute image, lost in an endless scroll of equally forgettable images, really worth sacrificing your privacy?
Read
Post-AI creative edge
In the midst of the AI boom not so long ago, those who jumped on it early stood out. Fast forward to today and it's no longer a novelty - everyone’s got it, everyone’s using it. But as Rob Schwartz writes for Print Magazine, once AI has given us all it can, it’s the creatives who will rise above the noise.
Read
Ital-AI-n newspaper
In a daring experiment, the Italian newspaper Il Foglio rolled out Il Foglio AI, a daily edition entirely generated by AI, with human journalists merely supervising. Unsurprisingly, the result is a shallow imitation of the paper's style, lacking any meaningful substance. Humans are vital to the loop, as we often say.
Read
AI lawsuits mapped out
News publishers and creatives worldwide are taking legal action over AI’s use of their work, with the NYT vs OpenAI saga grabbing most attention and NYT now demanding OpenAI hand over its profit documents. For those who prefer a visual breakdown, here’s a world map showing this and all other copyright lawsuits against AI companies to date.
Read
Success story of the week: Direkt36
Direkt36’s co-founder András Pethő walked away from Origo when turned it into Hungary's political megaphone. Now, he’s leading Direkt36, one of the few remaining independent voices, proving that independent journalism still has some fight left in it.
Read