June 19 2026 - Midjourney Spa, GPT-5.5 & AI Creative ToolsMidjourney Spa, GPT-5.5 & AI Creative Tools

**Taylor** (0:00)
Welcome back to another episode of AI Signal & Noise. It is Friday, and we have some absolutely wild stories to break down today.
I'm Taylor, and I am so excited for this lineup.

**Morgan** (0:15)
And I'm Morgan. Happy Friday, everyone. I hope you're ready because Taylor has been buzzing about today's news all morning. And frankly, I need some convincing on a few of these.

**Taylor** (0:26)
Dude, you have no idea. We're talking AI-powered medical upgrades, massive corporate shakeups, and, wait for it, an actual AI spa in San Francisco.

**Morgan** (0:39)
An AI spa?
Okay, now I'm actually intrigued, if not slightly terrified. Let's not keep everyone waiting. What is this spa business all about?

**Taylor** (0:50)
Okay, so you know Midjourney, right? The absolute kings of AI image generation. Well, according to The Decoder, they are building a full-body ultrasound scanner and opening a physical spa in San Francisco.

**Morgan** (1:05)
Wait, hold on. A spa? Like, with massages and cucumber water? And a medical ultrasound scanner?
Are you sure this isn't an early April Fool's joke, Taylor?

**Taylor** (1:17)
I swear it's real.
There have been rumors about Midjourney doing hardware for years, but nobody saw this coming. They're literally opening a physical space to house this high-tech scanner.

**Morgan** (1:32)
But why? They make software that generates pretty pictures. How do you go from digital fantasy art to scanning people's internal organs in a spa?
It seems like a massive leap.

**Taylor** (1:45)
Well, apparently, they want to use their advanced image processing tech to make ultrasound scans way more detailed. And like, beautiful?
It's all about advanced visualization of the human body.

**Morgan** (1:59)
I mean, medical imaging is incredibly complex.
If an AI hallucination happens on a digital painting, it's art. If it happens on an ultrasound, that's a massive medical liability, right?

**Taylor** (2:12)
Totally. And that's a super fair point. But they're aiming to make the whole experience less clinical and more relaxing. Like, imagine getting a health scan while getting pampered in a beautiful space.

**Morgan** (2:25)
Pampered by AI?
I don't know, Taylor. San Francisco tech bros will probably line up around the block, but I'll stick to a regular doctor and a normal massage, thanks.

**Taylor** (2:37)
Ha ha! Come on, live a little.
I would absolutely try it. Imagine the aesthetic of a midjourney designed spa. It's probably going to look like a sci-fi dream, very futuristic.

**Morgan** (2:51)
Or a sci-fi nightmare. But seriously, it's a fascinating pivot. It shows these AI startups are desperate to find physical, real-world applications for their tech beyond just web browsers.

**Taylor** (3:03)
Exactly. It's all about hardware integration. Anyway, speaking of medical stuff, OpenAI just dropped a massive health update of their own that is getting a lot of attention.

**Morgan** (3:14)
Oh, right. I saw this on the Decoder, too.
OpenAI is claiming their new model, GPT-55 Instant, actually beats real doctors at answering medical questions. What's the catch?

**Taylor** (3:28)
Yes. It's insane. They upgraded ChatGPT's health care capabilities, and in their own tests, the AI outscored actual doctors' inaccuracy, clarity and how complete the answers were.

**Morgan** (3:41)
Okay, let's pause. In their own tests is the key phrase there.
OpenAI testing their own product and saying it's better than human doctors makes me incredibly skeptical.

**Taylor** (3:54)
Sure, but look at the numbers. They said the error rate for health-related statements dropped by 71%.
That is a massive leap forward, dude. You can't ignore that.

**Morgan** (4:06)
A 71% reduction is impressive, yes, but a doctor has clinical experience, empathy, and most importantly, they actually know you as a patient. AI doesn't have that.

**Taylor** (4:20)
Oh, totally. Nobody is saying fire your doctor yet. But think about people who can't afford health care or just need quick, reliable triage at two in the morning. It's a huge deal.

**Morgan** (4:31)
True, but what happens when the AI gets that one critical thing wrong?
If a doctor makes a mistake, there's malpractice insurance. What happens if ChatGPT gives bad advice?

**Taylor** (4:44)
Well, they still have the usual disclaimers, of course. But if the accuracy is truly beating human benchmarks, it's going to save lives by catching things early. It's a net positive.

**Morgan** (4:56)
I hope so. I just worry people will treat this as an absolute replacement for medical care.
We need to keep humans in the loop, especially when it comes to health.

**Taylor** (5:06)
Agreed. It's a tool, not a replacement. But man, GPT-55 Instant is showing some serious power.
The future of medicine is coming fast.

**Morgan** (5:17)
It really is. But all this advanced AI is getting incredibly expensive to run, which brings us to our next story about Snap spinning off their team.