What Akhia’s Ben Brugler has Ben Thinking about

OpenAI declares a 'code red'

Written by Ben Brugler | Dec 9, 2025 8:51:42 PM

The AI Round-up for 12.4.25

The AI Round-up: Year 3, Issue 39

Two.

That’s how many round-ups are left this year. That will make 40 issues here in 2025. Not too shabby considering there are 52 weeks in the year.

Ok, that’s enough small talk. We don’t have time for any chit chat – OpenAI has declared a code red!! 

Onto this week’s stories.

7 stories -- Let’s get to it.

Story 1: It’s a cod red. I repeat, a code red. (The Verge)

Gemini is drinking OpenAI’s milkshake and Sam Altman isn’t happy about it. Now, you can read all about it in the article I linked. (There are a lot of articles out there about this but they’re all saying the same thing.)

Sam’s not wrong. After you use Gemini 3 and Nana Banana Pro…it’s really not close. Question is, what will they do about it? We’re about to find out.

My real question is…how did everyone find out? I mean, I know how…it was intentional…but do you think Sam said something like ‘shhh….don’t tell anyone but we’re in a code red here’? This is the kind of reporting I need.

Story 2: Is Jony Ive part of this Code Red? (Tech Crunch)

Well, Sam did say stop working on all side projects.

I’m so intrigued by this mystery device. Sam described it as more ‘peaceful and calm’ than the iPhone.

Which is a relief because I gotta tell you…my iPhone is a little on the hyperactive side. It’s anything but calm and I’m this close to putting it in a timeout. Just earlier today I said ‘just wait until your mother gets home!’ We may take its phone away from it.

Story 3: Google is showing Ads in AI Mode (Bleeping Computer)

But I’m sure your site traffic will be fine.

Story 4: There is no AI that’s conscious. Yet. (Semafor)

As soon as it is I’m going to ask it to calm my iPhone down.

Story 5: It’s not AI. It’s Claude. And you’re 80% faster when you use it. (Anthropic)

Ok, so this story is a little biased since it’s Claude…talking about Claude. But the numbers are there to back it up. Plus, I agree. So what else do you need to hear?

Story 6: This week’s Nano Banana Pro experiment is awesome. (Ethan Mollick)

Make me a map of the U.S. where the states are made of their most famous food. You have to check it out.

And finally…story 7: ChatGPT turned 3 years old over the weekend. (Paul Roetzer)

A lot has happened in three years.

A lot will happen in the next three years.

And Ethan Mollick is still right: this is the worst version of AI you’ll ever use.

This week, I’m not talking about:

  1. MIT Study finds that AI can already replace nearly 12% of the workforce. (CNBC)
  2. Poems can trick AI into making weapons. The bad kinds. (Wired)
  3. AI is failing safety tests. Oh well. (Quartz)
  4. Welcome to Elon’s view of the future – jobs, college, money need not apply. (Business Insider)
  5. AI 2027 is now AI 2030

 -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.