McClatchy Media runs newspapers around the US. But the management have fallen to the AI. Their exciting new plan is called “content scaling”. From Eric Nelson, vice-president of local news: [Wrap, archive]

Nelson was promoting the company’s new “content scaling agent,” an AI summarization tool powered by Anthropic’s Claude, which he said can help reporters find “new audiences, angles and entry points.”

“Journalists who embrace and experiment with this tool are going to win,” Nelson told the group, according to multiple people familiar with the meeting. “Journalists who are defiant will fall behind. Bottom line: We need more stories and we need more inventory.”

This is someone who thinks what their company does now is human slop, and they’re literally unable to tell if something’s actually any good.

It’s not clear who built the Content Scaling Agent or who sold it to McClatchy.

Journalists are now withholding their names from the bylines for the slop, under union and contract rules. [Wrap, archive]

But McClatchy is insisting on using the journalists’ names anyway, per Kathy Vetter, chief of staff for local news:

“We have every right to use their work,” she said, according to multiple sources familiar with the meeting. “It belongs to us, and if an editor wants to go … in there and repurpose a reporter’s content, they can put their name on it.”

Why would McClatchy want to abuse a journalist’s personal reputation like this? McClatchy’s blaming Google!

Nelson said the company needed to use reporters’ names to show “authority” with Google, whose changing algorithms have induced massive traffic plunges for news publishers, as it was “very important to say that this is from our reported stories.”

McClatchy is also trying to cover up that it’s publishing AI slop:

An April 7 update to the CSA [content scaling agent] said that the “discover explainer” drafts optimized for Google would no longer automatically include an AI disclaimer and that reporters must manually include them.

All this is for the Google that just screwed over everyone by trying to steal all their content and serve it with the AI.

There’s a union agreement at McClatchy against AI in the newsroom. McClatchy decided to just break that agreement, first thing! [Sacramento Bee]

Under The Sacramento Bee news employees’ latest contract, ratified in February, the Guild is supposed to receive 30 days’ notice before management deploys generative artificial intelligence technology in the newsroom, Lange said.

McClatchy rolled out the technology without telling them, she said. “It was a brand-new contract, and it was almost immediately violated.”

Reporters at several McClatchy papers went on strike over this.

This is the same trick that Politico tried to pull with AI in the newsroom last year. Breaking the union deal didn’t work out for Politico either. That union had a slam dunk win in arbitration in December. And just this week, Politico has shut down the AI newsroom tools: [NewsGuild]

We refused to back down, and POLITICO heard us loud and clear that these tools do not belong in our newsroom.

But I’m sure McClatchy will get there in due course.

• Video — Podcast

Pope Leo XIV presented his first encyclical letter to the Vatican on Monday. Titled “Magnifica Humanitas,” or Magnificent Humanity, the 43,000 word missive specifically takes aim at AI — and the AI bros. [Vatican]

It’s a book-length writeup on the AI bubble and why it’s terrible. It details all the current human abuses of AI in practice, and states that none of this should be allowed to happen. The Pope told the press: [Vatican News]

Artificial intelligence now demands to be disarmed, freed from logics that turn it into an instrument of domination, exclusion or death.

This is the Catholic Church going up against greedy technology tycoons and pointing out just how lethal and dangerous this technology is to the little guy, to democracy, and to the world at large.

It’s written by smart guys who have thought deeply on this for a long time. Pope Leo warns that the AI revolution is driven by the “idolatry of profit that sacrifices the weak.”

(Fun fact: the Pope’s undergraduate degree is in mathematics, with electives in Hebrew, Latin and theology.)

The Pope is well aware that the internet is filling up with AI slop. “The search for truth is an essential element of democracy,” he says, citing “Origins of Totalitarianism,” Hannah Arendt’ s 1951 book on the rise of Nazism. AI is producing the scenario Arendt pointed to as the prerequisite for totalitarian domination: destroying people’s ability to discern fact from fiction.

Pope Leo made his mission clear when he chose his papal name. It was in homage to Leo XIII, the 19th century pope who stood up for the common man against the industrial tycoons of the 1800s. AI represents the industrial revolution of the modern age with similar challenges.

So he is not just taking a stand against AI, the technology, but also the tech billionaires and their greed.

The timing of this missive is perfect, just as SpaceX, Anthropic, and OpenAI are getting ready to launch their ridiculous IPOs — though it’s unlikely the Pope cares. So far Anthropic have responded to the encyclical with vapid content-free sucking up. [Anthropic]

Pope Leo hates current AI so much. He’s been talking about the abuses of AI since he got the job in May 2025. This encyclical is how a pope expresses utter disgust with the AI bro creeps. It’s not just the AI tech, it’s everything it stands for and why the bros are doing all of this.

This encyclical will be the basis for what Leo represents moving forward. This will be Pope Leo’s battle and his legacy, and it’s the reason the Catholic Church chose an American to be the Pope

The encyclical says a lot of the obvious correct things about AI. Any Pivot to AI fan wouldn’t find a lot to disagree with. It’s nice to see someone with a platform this huge saying this obvious stuff.

So what does this mean? What follows from this?

An encyclical isn’t a set of religious directives. It’s a position paper. The Pope has not charged Catholics worldwide to burn down the data centres. Cool as that would be.

Catholics probably can’t go into work and claim a religious exemption from Claude Code. Though the encyclical does give Catholics who hate AI an excellent set of talking points to answer that AI bro who just doesn’t stop.

The encyclical does not have direct consequences. Functionally, this is just a letter. But it’s a letter that’s in every newspaper this week. It’s going to be influential.

The Pope calls for government regulation to stop AI abuses. And up against that, we have a ton of money. But the encyclical will still give politicians a bit of think tank input on things they have to consider politically.

Even the rich tech bros are treating this encyclical as a threat. The AI companies lobbied the Vatican quite hard in the leadup to the encyclical. We’re not sure they got a lot of what they wanted. [Politico]

But again, an encyclical is just a letter. Pope Francis did a quite good encyclical on climate change in 2015. Then he followed that up in 2023 annoyed that nothing much had been done. There’s only so much a letter, even from the Pope, can do in the face of the money.

This encyclical will help swing the vibe against AI, however. Maybe J.D. Vance will excommunicate the Pope. You know he wants to.

• Video — Podcast

OpenAI is running out of other people’s money. Time to hit the public markets! OpenAI hopes to IPO in September. [WSJ, archive]

SpaceX’s IPO is set for June 12. That’ll mess things up for OpenAI and suck a lot of the IPO money out of the broader market. [SEC]

The other problem is, even OpenAI’s own Chief Financial Officer says their accounts are trash.

OpenAI’s Q1 2026 revenue was $5.75 billion. But “operating income margin” was not great: [Information]

Adjusted operating income margin was -122% in the first quarter … That means that for every dollar of revenue the company generated, it lost $1.22 — even after excluding some large line items such as stock-based compensation.

“Adjusted” is an accounting jargon term meaning “made-up BS.” I suspect OpenAI’s losses were much larger.

OpenAI is trying for more revenue per user. They’ll need it — the actual user numbers have gone flat:

OpenAI’s weekly active users for the quarter averaged about 905 million users … Although the company hit about 920 million weekly active users in February, the lower average figure for the quarter implies that usage during the rest of the quarter was weaker than the February snapshot.

That 920 million figure works like an “annualised” number, where you pick your best week and you make out it’s typical.

OpenAI has stopped growing.

OpenAI’s next job is to try to skate past all of this, to the golden shores of the IPO! As long as these tawdry numbers don’t pull them down.

• Video — Podcast

Amazon still wants to go full corporate AI. Despite the February Amazon Web Services vibe-outage, which didn’t slow them down for a moment.

Amazon managers were wowed by OpenClaw. So Amazon made MeshClaw, which has been deployed across the company over the past several weeks.

Amazon has now given AI use targets to 80% of its developers. They even have a leader board in the office, to show who’s been maxing their tokens hardest.

So the engineers are using MeshClaw as much as they can: [FT, archive]

There is just so much pressure to use these tools. Some people are just using MeshClaw to maximise their token usage.

Amazon assured the devs their token statistics would not be used for performance reviews. The developers know better than to believe a word of it:

Several staff members said they believed managers were monitoring the data. “Managers are looking at it,” said another current employee. “When they track usage it creates perverse incentives and some people are very competitive about it.”

Amazon also lets MeshClaw do code deployments, because they learnt nothing from the February AI outages. Even as the employees point out how obviously stupid and risky this is:

The default security posture terrifies me. I’m not about to let it go off and just do its own thing.

Amazon still does not have an actual use case for AI. But that doesn’t stop your manager from setting a KPI and breathing down your neck. Use more tokens! Token it harder!

• Video — Podcast