Publishing has been defending for too long - it's time to go on the attack
It's always darkest before the dawn, and that means current predictions about the future of content are still informed by the dark. We think the dawn is imminent.
As the relatively narrow base of what GenAI actually does creaks ever more alarmingly under the weight of expectations loaded upon it, in the manner of an upside-down iceberg floating on its tip, there are more reasons to be optimistic about the future of original content and the publishers that still produce the vast bulk of it than has been the case in a while.
Interestingly, nothing has yet actually materially changed to cause this optimism, other than observing a technological timeline running at a pace that outstrips the reality of adoption, and, indeed, the willingness and sense to adopt. As we've pointed out previously, you can tell it's a gold rush when really the only business making actual money from it is the one selling shovels, that being Nvidia and their magical GPUs.
So as we see the very businesses that are involved in promoting GenAI, such as Meta and Alphabet, restrict the use of such content on their own platforms even as they are involved in an insane bidding war for the world's best computational brains to further the use of the same, anyone possessing even a wonky commercial compass must see the direction of travel doesn't quite add up. Particularly when the product so frequently fails to meet the standards of Production.
What has happened over the past couple of years leaves the publishing industry rather in the situation of the monks of Lindisfarne Abbey clearing up after a morning visit by some enthusiastic and heavily armed Viking tourists.
Notably, the Abbey outlived the Vikings.
There's no doubt we have all been pillaged for training data though, and many of us continue to be pillaged. It's more the result of what is being done with all the stolen treasure though that leads me to optimism. To continue the analogy, our content is being melted down and reforged by GenAI into something much much blander. Bland won't win. It must not.
Call me old-fashioned, but as a newsman, I prefer to base my reporting on deterministic rather than probabilistic methodology. Probabilistic doesn't do so well in court. GenAI is probabilistic all the way down. Once that simple realisation filters through to more people, attitudes will change.
Given that the GenAI maelstrom has at least given us some wild and absurd things to report on, it's not all bad. Other than the various legal and political avenues being pursued by original content creators against the data harvesters, the irrational GenAI scramble and its galactic spending war carries the seeds for its own end, without intervention. Given that is the case, what should publishers be thinking about now, in the darkest before dawn period? What does freedom from fear of Big Tech feel like?
A big step in removing such fear is the simple realisation that worrying about discoverability in AI search is largely a waste of energy. The rules of that game are so heavily loaded against publishers, and the reward window so pitifully small that "why bother?" is both the instinctive and logical position. We've even got a phrase for it - "Google Zero" - coined way back in 2024 and suddenly very real. Attempting to co-operate with someone trying to murder your business makes no sense. There is little point in expending resource on it.
Instead, that energy, freed from the uncertainty of trying to make Google like you this month, should be redirected to the core business, that of good content. Do what you do well and do more of it. Direct some zeal towards relationship building with your audience. It works. We are all both sides of the screen, producer and consumer. We know what works. Once loyalty is gained, it is only the work of fools to lose it. There's going to be an ocean of slop out there, if there already isn't, so be a safe harbour of quality content. Given how bad much GenAI content is, quality isn't even that high of a bar.
That also leads to having the most open mind about transactional relationships with your audience. Offers, contests, prizes, merchandise, exclusive access, direct access, whatever works and more. It's a slippery world out there, so that relationship of trust can yield financial benefits to publishers. Do an audit of the direct and indirect relationships you have, and look for them in unexpected places, they can yield benefits to both parties.
What has come to be called churnalism might be in trouble, at least at the lowest and most numerous end of it. GenAI can well cope with turning round a formatted press release or introducing stats to an accepted template. but other content is not reliably to a high standard though, speaking as someone whose journalistic career saw me spin some very rough thread into content gold, or at least gold-plated. Humans are good at stuff they know how to do well. Spend money on them.
Being slow to adopt or adapt is fatal. In some respects, if we are facing the death of search, and that means all previous ways of thinking about visibility are obsolete, then a tremendous freedom is being offered to the industry. We have been playing the second guess game so long with the Big Tech that sometimes it seems we've forgotten how to guess for ourselves. That has to change, and organisational agility in tech and people is vital.
This is harder for some publishers to swallow than others, particularly for those not big enough to be big, but not small enough to be small. We are exiting an abusive relationship, and what happens next it up to us.
It's my personal belief we are on the verge of an age in which the smaller, relatable, controllable and more personal spaces will come to be valued more highly once again. People have a well-developed suspicion of the huge, and watching mega corporations roll across the planet on a suffocating carpet of seemingly endless cash is getting tiring.
You have nothing to lose but your soon-to-be non-existent SEO ranking, so get up and get at them. As Marshal Foch said: "Hard pressed on my right. My centre is yielding. Impossible to manoeuvre. Situation excellent. I am attacking."
Calling all media and publishing content leaders, product leads, data strategists and tech innovators - we want to hear from you!
We’ve teamed up with State of Digital Publishing to launch a benchmark study assessing first-party personalisation, CDP adoption, and audience engagement in digital media and publishing.
This report will:
Benchmark CDP adoption and performance
Reveal emerging personalisation trends
Identify challenges and ‘Aha’ moments from across the industry
Deliver actionable best practices
Give us 5 minutes to complete the survey and we’ll give you early access to the findings to see how you compare.
AI bot's Perp walk
Shock! Perplexity is accused of scraping content from sites which explicitly said "No AIs allowed!" and had set bot protocols and firewalls to block the firm. Bot foe Cloudflare says AI snoops used the old Scooby Doo ploy to bypass site protections to keep the content funnels flowing despite de-authorisation, naming and shaming Perplexity. Perplexity, as happy to snoop its own users as it does other people's content, dismisses it all by basically branding Cloudflare a doddery granddad who doesn't know how the web works. Maybe they wrote that bit themselves. Would such a scenario if true also risk Perplexity inadvertently missing out on making payments under its vaunted Publishers Program set up to woo content owners? Let's hope not.
Read
OpenAI's sharing "slip"
Speaking of data sharing, ChatGPT/OpenAI managed to send troves of user chats to search engines via a new share feature - perhaps hoping lots of new users might flood the service to force positive chat statements about their company into Google results. Only issue was, the warning text that private chats could be sent to the world was so obscurely and faintly placed that all sorts of sensitive info was scooped up without anyone realising. It's all OK now though, says someone with a cool X handle - OpenAI CISO DANΞ says the company pulled the feature and is now somehow cleaning up Google and other search engines. He was formerly CISO at Palantir so can perhaps be excused for the novelty of removing data.
Read
AI giants face EU's new rules
In an era when AI is evolving faster than regulators can blink, the EU has drawn a line in the sand with its AI code of practice, to which tech firms broadcasting their intentions can sign up. Now we know who is in, and who's out, the likes of Amazon, Google, Microsoft, OpenAI, Anthropic and even Grok have pledged to play by rules on transparency, safety, and copyright. Not so Meta, which basically gave the bird to even looking like it thinks rules are worth reading. With the AI Act closing in, those who skip the code might not have a good time when it finally comes around.
Read
Hype meets regulations
Ludicrous AI marketing claims are under the hammer after regulators started demanding honesty and clarity in the marketing of AI capabilities. With initiatives such as the FTC's "Operation AI Comply", the message is clear: transparency is not optional, but mandatory.
Read
Summer SEO lessons
This summer, semantic SEO is making a quiet comeback in giving pages and content real structure and meaning in a world where machines still need to make sense of your pages. SEO magician Barry Adams discusses how semantic markup is not only good practice, but also a competitive edge.
Read
ONA's open AI conundrum
Unsure how best to flag the use of AI in content to audiences? Or even if you should? The Online News Association compared different ways of presenting AI usage declarations, and their impact on trust. There's big variations in results but notable is that the mere mention of AI reduces trust - even when that mention says "AI was not used". ONA is the world's biggest digital journalism organisation and the research webinar is free and very useful. We'll be at ONA25 in New Orleans next month, so do come and say hi.
Watch
Make the most of your beach bar break
Looking for work reading on your summer hols to make the bar bill tax deductible? Our friends PugPig have you covered with their 2025 Media App Report giving a front row seat to what's really working in mobile media today. The PugPig team gathered insight from 140 publishers and 400 apps, looking at AI experiments to the rise of audio, new App Store rules and the power of newsletters. It's a cheat sheet for any publisher app strategy.
Read
Free online safety speech
After a clash with French prosecutors, Elon Musk's X is now crying foul play over the UK's Online Safety Act. X claims the law is about censorship not child safety, with strict enforcement and mega-fines for anyone who doesn't comply. The clash has sparked political fireworks, turning a safety law into a full-blown opinion shutdown.
Read
CPB signs off after 60 years
A non-profit which helps fund more than 1,500 local radio stations across the US is shutting down after nearly six decades, following its funding being pulled. The Corporation for Public Broadcasting is winding down after Congress U-turned on funding commitments at President Trump's behest. The CPB was a lifeline for many independent rural outlets, many of which are expected to shut down as a result. Poynter shares more details.
Read
The unexpected powerhouse fuelling AI models
Reddit has surged ahead as the top source for AI models due to its vast stores of user-generated content. While ChatGPT has a preference for Wikipedia, other AIs rely heavily on Reddit's human conversations to learn. Press Gazette has the scoop.
Read
Another must read by WAN-IFRA
What do you get when you gather 120 news leaders from around the world and ask them how they're navigating the chaos of modern media? You get a playbook, and that's exactly what WAN-IFRA did. After this year's WAN-IFRA Congress in Krakow, they released a practical playbook packed with ideas and smart strategies into key areas which shape the future of news.
Members read here
YouTube's AI test raises privacy concerns
YouTube is rolling out AI-powered age checks in the US, attempting to guess who's under 18 based on viewing habits and search patterns. Will YouTube and Google endeavour to make the system work well? Hmm. Over-flagging over 18s as "unknown" and thus needing to hand over their ID, would be a massive win for Google should it plan to become the dominant verifier of IDs in the future. Privacy experts are not a fan, but probably nor are those running all the fake sites popping up to "check your ID" as government Big Brains normalise sending your biometric details to who knows where. Ironically, the under 18s are probably best at spotting the scam sites - if only they could tell us!
Read
SEO's new frontier
So what of the “legit” web traffic coming to sites via AI - regular users just trying to find things. Here, AI is showing gains, with a significant 527% surge in AI-driven visits early in 2025. ChatGPT, Perplexity, and Gemini are quietly steering users to digital doorsteps, especially when it comes to legal, finance, and health sites. Charming Google is still the top priority, but effort to woo AI assistants and chatbots is starting to show some reward. Search Engine Land knows more.
Read
AI joins the newsroom, but with a human editor on speed dial
AI is clocking into newsrooms, but they still need oversight from a grumpy editor. Tools powered by RAG (retrieval-augmented generation) are speeding up tasks such as data analysis but while these bots are helpful, as you might expect they still hallucinate, oversimplify, and fumble details. Both the newsrooms loving the help, and the sceptics saying they do more harm than good, can agree on one thing: AI needs an editor.
Read