Why Jev refuses to write sentences

SPEAKER_1 (0:00)

When you need to build up your team to handle the growing chaos at work, use Indeed Sponsored Jobs. It gives your job post the boost it needs to be seen and helps reach people with the right skills, certifications and more. Spend less time searching and more time actually interviewing candidates who check all your boxes. Listeners of this show will get a $75 sponsored job credit at indeed.com/podcast. That's indeed.com/podcast. Terms and conditions apply. Need a hiring hero? This is a job for Indeed Sponsored Jobs.

SPEAKER_2 (0:31)

Low energy, unfocused, foggy, you might be dehydrated.

Gatorade Zero powders have been scientifically designed to help improve hydration balance in the body. With a clinically proven electrolyte blend and no artificial colors or flavors. Great tasting hydration no matter where you're going. Shop now online or at retailers nationwide. Gatorade hydrates better than water. Is it in you? Symptoms noted are signs of mild to moderate dehydration. Consult a health care professional if symptoms persist.

SPEAKER_3 (1:00)

Your heart can tell you a lot about your health. Apple Watch Series 12 measures your heart rate every five seconds with the most accurate heart rate sensing in a wearable. So your vitals app, now with heart rate variability, can tell you when something is off. And your readiness score can let you know when to rest and when to push. Hear the story in every heartbeat with Apple Watch Series 12

The features described are for wellness purposes only and not for medical use. iPhone 11 are later required. Based on Apple conducted study of heart rate accuracy August 2026, visit apple.com/applewatchseries12.

Stage Zero (1:30)

A former OpenAI researcher named Diogo Almeida just built an AI model called Jev for his startup TypeSafe AI.

And the financial press is floating valuations in the billions. But the weirdest part isn't the money. It's that Jev explicitly refuses to write a single sentence.

SPEAKER_5 (1:46)

I'm right, which goes entirely against the current grain of everything happening in the tech industry right now.

Because the entire space is just obsessed with chatbots and generative text. Every product, every interface, and every backend system is being rebuilt to have this fluid, open-ended conversation with you. Yet this specific model operates by just selecting from preset responses and spitting out a strict confidence score. It acts completely counter to the prevailing logic of what an artificial intelligence product is actually supposed to look like today.

Stage Zero (2:19)

So, why are developers suddenly abandoning conversational language models to use a bot that acts like strict triageners?

SPEAKER_5 (2:27)

Well, it really comes down to the underlying economics of what we are actually asking these systems to do in production environments. When you look at real-world applications, generating free-flowing text is often the exact opposite of what a developer actually needs for a backend system.

We have been applying this conversational tool to mechanical sorting problems, and developers are realizing that forcing a conversational AI to do administrative work is computationally expensive and frankly, practically inefficient.

Stage Zero (2:58)

Yeah, and you can actually see the friction of that mismatch in the performance data because Polylane swapped their traditional language models out for Jev in their production environment and found it was 89% faster and 39% cheaper.

SPEAKER_5 (3:10)

Oh, wow. And Sam Rome looked at a series of classifier tasks and found an even wider gap. Depending on the workload, moving away from a traditional large language model made the process up to 400 times cheaper and 200 times faster.

Stage Zero (3:24)

Right. And Prithvi Datla really crystallized why this gap exists by pointing out a very simple reality of the technology, which is that reading is cheap, but writing is what you pay for.

SPEAKER_5 (3:34)

Exactly. We have been forcing language models to write conversational essays when we really just needed them to sort data.

Because generating words requires the model to calculate the probability of the next word token by token over and over again. Every single time it outputs a word, it scans its neural network. It is mathematically exhausting. If you just need a system to verify a condition with a yes or a no, having the model generate a conversational paragraph explaining its reasoning is entirely wasted compute. You're just burning money on words you don't need.

Stage Zero (4:08)

You really have to look at how Datla tested this concept on Texas government contracts to fully grasp the scale of the waste. He looks specifically at requests for proposals.

SPEAKER_5 (4:17)

Right. Which are incredibly dense bureaucratic documents.

Stage Zero (4:21)

Completely.

They are filled with legal stipulations, compliance requirements and specific deadlines. He mapped out 438 complex rules to see how different systems handle the reading and sorting process. He was essentially simulating the grueling administrative that usually requires a small army of paralegals or compliance officers to sift through line by line.