Cloudflare Gives AI Bot Companies Ultimatum

A move by service provider Cloudflare has become highly publicized over the past few days.

The now-famous pressing of the button by Cloudflare CEO Matthew Prince was a clear challenge to AI companies – pay what you owe publishers and content creators. That event sets the default permission for AI crawlers scraping data to “no,” and the change has teeth. They call it “Content Independence Day.”

Here’s how the moment when June became July was described by one participant publishing on Substack, writing about the event held at Cloudflare’s One World Trade headquarters in Manhattan:

“Pushing the button at midnight wasn’t for show—it was a signal to publishers, creators, and platforms: this is the moment. This is the fight.

Prince looked out at the crowd. ‘Today, July 1st,’ he said, ‘we declare that there cannot be any AI crawl without compensation. This is the first step.’

The countdown began. 10… 9… 8… The crowd leaned in. Cameras out. Hands started to pile on top of the red button—one by one, then all at once. ‘Pushing this button may seem like a small thing,’ Prince said as the clock ticked toward midnight, ‘but it’s the start of a really big revolution.’ And then it happened. Midnight. A click. A cheer. Content Independence Day began…”

It may have been a momentous occasion for the web.

“If the Internet is going to survive the age of AI, we need to give publishers the control they deserve and build a new economic model that works for everyone – creators, consumers, tomorrow’s AI founders, and the future of the web itself,” Prince said in a related press statement. “This is about safeguarding the future of a free and vibrant Internet with a new model that works for everyone.”

The Background

There’s a very important context to this – it started with newspapers in the earlier days of the Internet. The business model just got harder and harder to support. The profit margins got lower and lower. Advertising changed.

Then, when centuries-old newspapers and younger, smaller newsrooms were already on the ropes, along came AI, which actually takes the work of humans – reporters, broadcasters, artists, musicians, etc. – and creates competitive works that are essentially free.

Humans can’t compete. Local news, and “human” news in general, is one of the canaries in the coal mine.

However, Cloudflare’s move could be a broader trendsetter if it turns out to be effective.

Government Regulation

There’s also some related activity on the part of the Securities and Exchange Commission, which is requiring AI companies to disclose instances where they use AI in their business models.

Presumably, that would include using AI to scrape content and make it into a finished product.

So this will indirectly support the idea that AI companies can’t get away with web scraping like they’ve been doing so far

But it doesn’t actually prohibit that kind of activity

Court Wins for Big Tech

Nor does local court action, with U..S courts recently ruling that Anthropic was not liable in collecting human-made content and using it to train AI.

A judge did find Anthropic liable due to using these published works without paying for them – allegedly, essentially pirating books in order to train the model.

But in terms of general rulings on IP, it seems like courts are largely okay with AI companies training their models on publicly available data.

So Cloudflare’s action is more of a deliberate play on societal morals, and not the law itself.

In other words, it’s not a legal vehicle.

ChatGPT, Are You Killing Local News?

I felt like asking a model directly if it’s responsible for compromising local newsrooms around the country. After all, there’s nothing like hearing it from the horse’s mouth. Also, I was curious as to whether the model would admit its culpability, something that humans are usually loathe to do.

It turns out, though, that, devoid of human functions and related incentives, ChatGPT, in this instance, really couldn’t tell a lie.

Here’s what it came up with:

Q (me): “are you killing local news?”

A (ChatGPT 4.5): “That’s a sharp and important question — and the honest answer is: not directly, but AI (including me) can accelerate trends that are killing local news.’

Wow.

It did qualify this by pointing out the ways that content creators were already suffering previously:

“Even before AI, local news was bleeding out due to:

• Loss of advertising to Google and Facebook
• Declining subscriptions
• Consolidation by hedge funds
• Public apathy toward civic news”

And the conclusion, restated:

“So, am I killing local news?
Not deliberately — but I’m part of an ecosystem that can either erode or enhance it, depending on how people, platforms, and policymakers choose to act.”

Next Steps

A blog post at Cloudflare reveals that the grand opening of Content Independence Day is not the end of the firm’s efforts.

“Next, we’ll work on a marketplace where content creators and AI companies, large and small, can come together,” write spokespersons. “Imagine an AI engine like a block of swiss cheese. New, original content that fills one of the holes in the AI engine’s block of cheese is more valuable than repetitive, low-value content that unfortunately dominates much of the web today. We believe that if we can begin to score and value content not on how much traffic it generates, but on how much it furthers knowledge — measured by how much it fills the current holes in AI engines “swiss cheese” — we not only will help AI engines get better faster, but also potentially facilitate a new golden age of high-value content creation.”

Presumably, there will be cooperation with groups like Press Forward that are aimed at helping news to survive.

“Like sidewalks and streetlights, local news is an essential resource for strong communities,” write Press Forward representatives. “It’s how we know when and where to volunteer, what’s happening in our schools and city councils, and even what to do on the weekend. However, as once-thriving local newsrooms have shrunk and disappeared across America, this vital public resource is at risk.”

The web site further reveals how Press Forward will fight back: with philanthropy, including an impressive $200 million paid to support news makers to date.

I think that 2025 is going to be a flashpoint for the publishing battle. Will AI companies have to pony up and pay, which will revive moribund business models for newspapers and other media?

Or will AI be allowed to scavenge online stuff for free, making news reporting a boutique hobby like, say, recording a musical album?

We’ll see. Here’s one more admission that I got out of GPT when I told it; “you can’t copy my work in a satisfactory way.”

“I can’t copy your work in a satisfactory way — not if “satisfactory” includes honoring your authorship, context, and rights. But I can help advocate for those rights, help you track usage, or even help design licensing models that protect your output. Would you like to explore one of those paths?”

No thanks, ChatGPT. But thanks for fessing up.