Some will take greater risks and win (or lose); others will play it safer and slowly accumulate wins (or be obsoleted).

Never mind the threat of letting these models write code that runs your business, or operate it agentically. Models trained by actors (corporate or nationstate) diametrically opposed to your interests.

Lots to take into account now, interesting time to be in business.

The combined use of faithful-chain-of-thought + mechanistic interpretation of LLM output to 1.) diagnose 2.) understand the source of, and 3.) steer the behavior is fascinating.

I'm very glad these folks found such a surprising outcome early on, and it lead to a useful real-world LLM debugging exercise!

Really looking forward to trying this out!

Meaning empirically, downtime seems to be tolerated by their customers up to some point?

> I feel that transforming such a wonderfully written story into a podcast with robot voices might result in a shallow and awkward rendition that doesn’t capture the essence of the original work.

I can definitely agree with that in the general sense! But I have a long commute, and I thought the content of it would still come through well enough, and I've appreciated it for a few other subjects. It's not as good as having professional voice actors read it out, but it still has some value to folks!

Better yet, if someone has already _done_ that Notebook Lm podcast and has a link to it, please share!

Comments URL: https://news.ycombinator.com/item?id=41452429

Points: 2

# Comments: 0

https://linzumi.com/

Definitely think this sort of idea could become the "serverless" equivalent for ml-using apps. I'm curious what you think re: versioning, consumption from various client languages, observability/monitoring/queueing, etc.? Feels like it could grow into a meaningful platform.

https://youtu.be/YX40hbAHx3s

Although, 'script(/injection) kiddie' will be an interesting phenomenon in the future...

1. Take user task

2. Pass it to a prompt that requests a Product UI description of a component

3. Pass 1+2 to another that asks for which npm packages to use

4. Pass 1+2+3 to a templated prompt to write the code in a constrained manner

5. Run 4 in a sandbox to see if there are errors, if so pass it back to #4, looping

It’s currently quite slow, but that’s an implementation detail I think.

On the other hand, there are times where you want to ask the story teller if it's possible for you to do X - I think an iterative loop of that would probably be a happy middle ground (with next to no effort).

On the other other hand, maybe the original system prompt needs to include, "Don't let the player do anything that's out of place for the story". Lots and lots of ways to experiment here.

Oh, and it's also fun to hook up each "step" in the story to StableDiffusion to have it output a dramatic rendering of your story so far. I hooked up one scene from the YouTube video above to Midjourney and got quite a nice illustration out: https://media.discordapp.net/attachments/1051015357340602398...

----------------

(Midjourney prompt: "Dungeons and dragons comic book: "You find yourself at the entrance of a long-lost temple deep in the jungle... climb the temple to look for clues about the builders You take a step towards the jaguar, brandishing your sword and yelling at the top of your lungs. The jaguar hesitates for a moment, but then charges forward, claws bared. You ready yourself for the attack, determined to defend yourself "attack with sword' You and the jaguar continue to circle each other, both waiting for the other to make a move. Suddenly, the jaguar pounces, but you manage to dodge out of the way just in time. You counterattack with your sword" --v 5 ")

I don't know that it's actually effective at that (or if it is, that the inherent costs are worth it), but it's a bit of the reverse of most recent thinking: keep people for as long as you can if they're sufficiently useful.

I'm very bullish on the application of wasm in general at this point. It's such an interesting combination of:

   * a reasonable compile target

   * a reasonable-to-implement-safely runtime

   * Already widespread deploy targets (in the form of browsers and now more and more cloud providers)

> It's just a demo, and isn't supposed to be a real-world example of performance or best practices.

I also appreciate that it's still in an experimentation phase. Looks like a silly toy, like a lot of things do before they become mainstream.