The AI Round-up: Good? Bad? Does it matter?

AI news covering the week of June 22, 2026

(covering the week of June 22, 2026)

Frog.

Kettle.

Yada, yada, yada.

Boiling.

The discussion/debate over AI continues. Will it? Is it? Won't it?

I'd argue none of IT matters only because whatever IT is we've been doing for 3.5 years now...which makes me think the real question should be 'if AI went away tomorrow would you miss it?'

Specifically, your LLMs? The things you've grown used to doing on ChatGPT? The work you collaborate on with Claude?

I'm not sure if it's a question of 'do we want to'...as much as it's become 'can we'? A lot of people can't stand social media. Yet...well, you get it.

In light of how you answer that question...reflect on the fact you developed this opinion in less than four years. Has anything ever become so polarizing, so quickly?

Of course, LLMs won't be going away so the answer may seem pointless to you. But I felt the need to shift the perspective a bit because while we think we're in one of two camps...we're all in the same pot.

Let’s get to it.

Tag! Claude’s it.

Introducing Claude Tag – where you can ‘tag’ Claude in Slack (@Claude) to do…well, a lot.

I have news for ya…Anthropic won’t be the first company to offer this. And Slack won’t be the last app to integrate it.

RIP LLMs?

I’d read the linked content because it goes to Ethan Mollick who is sharing his thoughts on it as well as what it means for company governance.

So, is Anthropic on the naughty or nice list – which is it!?!

Despite creating Mythos/Fable, which was shut down by the government…the government no longer sees Anthropic as a security threat.

Apparently they’ve learned to be more ‘respon$ible’.

But wait…Claude/Mythos just broke into the government’s most classified documents in hours.

It’s ok, it’s ok…this is all part of Project Glasswing. And none of this sounds like science fiction.

OpenAI gets the Anthropic treatment.

Despite not wanting to limit AI development, the Trump administration has now slowed the release of a second company’s new, dangerously powerful AI.

OpenAI has been ‘asked’ to limit the release of its new model to government users only.

So, let’s recap:

=No guardrails for AI – winter 2025

=Slow down! – spring 2026

=Everyone and everything is getting too expensive – summer 2026

=China says…’well hello there!’

Are we sending users to Chinese-backed AI due to costs…which will help their AI technology exceed the United States’?

Yes. And if you listen (and believe, which you should) what Paul Roetzer says in this week’s episode of The Artificial Intelligence Show (#221), a ban on Chinese AI may not be far off…

We got here, fast.

OpenAI introduces its partner program.

Your Innie will enter your organization and train everyone how to use AI and build the whole operation around OpenAI’s offering. Your outie will never know.

I’m kidding. About the innie/outie part.

Consulting companies are being trained to help you understand (Open)AI so you can leverage that technology for your organization. No word on if you’ll still need said consultant but they work for Bain, Accenture, etc.

Let’s get physical!

According to Goldman bankers, the next AI boom is in the ‘physical economy’, i.e. the resources needed to build and give us more AI.

Good news! Your employees are using AI. Bad news…they aren’t telling you.

Ever since I’ve been tracking AI trends/news, the question of ‘who owns the AI’ has come up. Specifically, if an employee uses AI to make something, do they own it or the employer?

The answer I keep coming back to is ‘were they at work, on your time, on your license, etc.’ when the employee developed the AI tool/workflow/process? If yes, I guess the employer owns it.

Hence the reason 57% of employees are hiding their use of AI at work. (HBR)

It’s all starting to blur together.

Getty Images has announced a partnership with OpenAI, allowing users access to Getty’s image libraries.

That’s a pretty savvy move by Getty (the stock market agrees) to stay relevant. And an even smarter one by OpenAI to gain access to images to train models and just look a little more ‘human’ at a time when they need some good PR.

Of course, they didn’t say any of that but…come on. I mean…

I’m starting to wonder who will play Sam Altman in the movie that will no doubt be made about all of this.

Some Nopes.

The 5 eyes have spoken. Danger is months, not years away. (The Guardian)
Are Deep Fakes real…if you think they are? (The New York Times)
Humans understanding can’t keep up with AI advancement. (Tech Xplore)
Facebook now has an AI genie to answer your questions, like Google Overview. Difference is, the answers are pulled from posts and reels. So don’t ask. (DigitalTrends)
The G7’s AI coalition is the cartel now. (Shelly Palmer)

-Ben

As a reminder, this is a round-up of the biggest stories, often hitting multiple newsletters I receive/review. The sources are many … which I’m happy to read on your behalf. Let me know if there’s one you’d like me to track or have questions about a topic you’re not seeing here.

-Ben