dkersten

You could make it a requirement only for people who want to attribute “made by”. If you publish a no name extension, you could skip it.

Can you share a sample prompt demonstrating it? I’d be interested to see, especially since you’re saying it’s incredibly easy (20 seconds would mean it’s easier than on most normal LLMs even).

It separates context/state from prompt/questions. That’s a key ingredient for prompt injection hardening. If it is designed and trained to never treat state as instruction, and you only ever put untrusted user input into state, and never the questions, then why wouldn’t it be?

The jev documentation says this is what you should do. It doesn’t mean that it’s actually designed for prompt injection resistance, but it could be. We won’t know for sure until they either release more details or someone proves otherwise.

But the split is one that doesn’t exist in normal LLMs and it’s the exact split that is needed for immunity or resistance to prompt injection.

I’m not claiming you can’t get some of the properties in some ways with LLMs, just that Jev is the sum of its parts, not just one or two properties.

LLMs can produce structured output by limiting the next token based on a grammar, that ensures correctly formed output.

But (presumably, again I haven’t seen Jev’s insides) Jev doesn’t need to do that, it just has to output probabilities for each answer, and can do it natively without artificially limiting output tokens.

So jev is cheap, fast, never produces malformed output, its output always carries correct semantic meaning (but this doesn’t mean it always answers correctly), and it separates correct from prompt (which if it truly does that is the most exciting part).

My point is that it’s not replicated. You replicate the accuracy, but not the other properties. The Jev-competitors only proved that you can get or beat the accuracy, nothing about the other properties. Especially the “zero hallucination” output and the (if it works how the documentation make it sound) prompt injection resistant architecture. You can’t get that with a fine tuned LLM.

Most of them appear to be small LLM’s fine tuned for the role.

That’s a different set of properties in terms of size, cost, and latency. Jev (apparently, not like I’ve seen its insides) is extremely cheap, extremely fast, doesn’t cost any output tokens as it speaks the output natively, can’t get the output wrong because it speaks the format natively, and (presumably based on the docs), the context is separate from the question, meaning it should be immune (or at least highly resistant) to prompt injection attacks.

It’s not just about the accuracy of the result, it’s a collection of all the properties that make Jev interesting.

Jev took years to develop, I strongly doubt that a copycat that was put together within days after Jev’s release will be able to match it on a sun of its properties. Even if fine tuned LLMs can outperform it on raw accuracy.

So "Deno" isn't joining Cloudflare, Deno is effectively defunkt and its former team is joining Cloudflare.

Well, yes. Amusing might have been a better word than funny.

> if minor data or chat sessions are mishandled without explicit parental consent

I know this isn't what you meant, but I can't help but laugh at the thought of a parent consenting: "I consent to the mishandling of my child's data"

As to your actual point, yes, I assume they are trying to reduce liability, especially if they're gearing up to an IPO.

Interestingly, for me on iOS, all of the examples worked except the textarea one, which didn’t allow me to interact at all.

It’s also glitchy and flickery for me on iOS.

I enjoyed this talk. It’s long, but it’s interesting and goes into a lot of “lost” history.

Exactly, not all tasks require smarts, they just need to be good enough. I’ve found that if prompts are really focused, the tiny models do alright.