Skip to content
Gemini 3.8 Live and Jev Are Shaking Things Up artwork

Gemini 3.8 Live and Jev Are Shaking Things Up

The Daily AI Show

September 16, 2026

The episode focused on a shift from AI as something people prompt to AI as a system that continuously sees, listens, decides and routes work while people are using it. Gemini 3.8 Live provided the clearest example.

Speakers Brian, Andy, Gareth

TopicsTechnology

Brian (0:00)

Hey, morning, everybody. Welcome to The Daily AI Show. Today is September 16th, 2026, and we are live, as always, at 10 a.m. Eastern, 7 a.m. on the Pacific Coast where Andy is. Andy's here. I'm Brian. I'm on the East Coast. So we have the country covered at the moment. We think we'll have Gareth coming in maybe a little bit. Beth is under the weather, so she unfortunately will not be with us today, but I'm sure she will have a speedy recovery and get back to us as soon as she's able to. Maybe we'll see her in the comments, we'll see how she's doing. But yeah, Andy, I know the last two days, man, it's been longer conversations, and I figured today was probably a great day to get back to some of the other news stories that we haven't either just not had time or they were towards the push of it. And I wanted to share one that, it's kind of like multiple stories, but it's all about IFA, or if they say IFA, I'm not sure, I think it's IFA 2026, which is in Berlin every year. And a lot of people see this tech conference and sort of trade show as being the precursor to what you'll see at CES in January in Vegas. So it's a great precursor, like, okay, what are we doing? What are the robots? What are the things we're slamming AI into? It's always fun in January when we have the CES conference and stuff like that. But this is perhaps some of the same ones. Now, the two guys that happened to do this, Cypernews and Wearable are the two, they were doing a video together. As you can imagine, Wearable, that guy was all focused on like horror rings or variants of those or whatever. The Cybernews guy was pick out of anything. By far though, what I saw as at least a trend on their video though, Andy, was that it was a lot of AI focused on children, which you would imagine has some areas of concern, especially when it comes to video and anything with a camera on it.

I'm also curious too why more of these types of tools don't do things that are just with infrared or other types that don't need an actual full camera, but can still figure out movement and heat signatures. I just feel like that's such an easier path in if you're trying to go this route. But I'm going to bring one up on stage that I can't figure out for the life of me, because they definitely said this was for children.

I'm just going to play a little bit of it here for anybody who's not watching. It's a very creepy, as usual, humanoid looking robot, I guess you will.

Yeah, he's got a bubble beyond a shirt. Anyway, I'm just going to play it. You can hear a little bit of the guy talking about it, what the guy who works there is company saying what they're planning to do here, because I found this interesting.

SPEAKER_2 (2:43)

Behind me is an early entrant into perhaps the creepiest products of the century. It's Heart With Mind Robotics Child Form Silicon Robot.

SPEAKER_3 (2:53)

We call it Duncan because we tried to build this robot as much as it looked like a five years old percussion child, a boy. Originally, this robot was built for special children, for example, the neurodivergent children.

SPEAKER_2 (3:08)

Can we get a couple of the demonstrations of the things Duncan is capable of doing?

SPEAKER_3 (3:12)

Yunfeng, Yunfeng, who are you?

SPEAKER_2 (3:16)

I am Duncan, a childlike robot.

Brian (3:19)

I'm just an neurodivergent children.

SPEAKER_2 (3:21)

If somebody smiles at him, he can smile back.

SPEAKER_3 (3:25)

See? Electronize, it'll break it, and your view is bumped.

SPEAKER_2 (3:32)

Okay. Oh, man.

Brian (3:37)

All right. So that's the part I want to talk about.

Andy (3:38)

This is painful.

Brian (3:39)

It's painful. So anybody who's not watching is just listening. I know you can't see what's on the screen. But first of all, it's like not great, right? You heard the voice. It's like, I am Duncan, right? It's just like sort of robotic voice or whatever. The face is sort of creepy. Yes, it has some motors behind the silicone skin or whatever. But then it goes to raise its hand to do what I assume is a handshake, and it's this really janky, like bumpy, and that's what Andy was just laughing at too. Like he starts to raise a hand, and honestly, Andy, I could let it keep playing. It never gets better.

Andy (4:13)

It looks like Frankenstein.

49 more minutes of transcript below

Thousands of transcripts fetched by people building searchable podcast archives

Fetch the whole transcript

The demo key returns a sample episode in full, no card needed:

request
curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Markdown with the speakers named, for your notes, your knowledge base, or anything that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.

Using your own key:

request
curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/1000790091740