Hacker News Reader

Comment by Imnimo | Hacker News Reader

I'm curious what the applications are where people need to generate hundreds or thousands of these images. I like making Ghibli-esque versions of family photos as much as the next person, but I don't need to make them in volume. As far as I can recall, every time I've used image generation, it's been one-off things that I'm happy to do in the ChatGPT UI.

View on HN

As usual for AI startups nowadays, using this API you can create a downstream wrapper for image generation with bespoke prompts.

A pro/con of the multimodal image generation approach (with an actually good text encoder) is that it rewards intense prompt engineering moreso than others, and if there is a use case that can generate more than $0.17/image in revenue, that's positive marginal profit.

An obvious one is for video games, interactive fiction, that sort of thing. AI dungeon with visuals could be pretty interesting.

It's too expensive for that unless you had a pretty generous subscription fee. I think local models are probably best suited for gaming where a decent GPU is already likely present.

I think there is a niche for both. Local LLMs are orders of magnitude smaller, so you could imagine cloud bursting for the difficult/important work like generating character portraits.

That said it’ll be 10-20x cheaper in a year at which point I don’t think you care about price for this workflow in 2D games.

I use the api because i don’t use chatgpt enough to justify the cost of their UI offering.

I’ve built a daily image-based puzzle that’s fully automated, and have been using flux to generate images. I’ve found sometimes they’re just not good enough, so have been doing some manual curation. But, with this new API, I’ll see if it can run by itself again.

AI-assisted education is promising.

I'm still struggling to see how you would need thousands of AI generated images rather than just using existing real images for education.

The company I work for generates thousands of these each week for children's personalized storybooks to help them learn how to read. The story text is the core part of the application, but the personalized images are what make them engaging.

- personalization (style, analogy to known concepts)

- specificity (a diagram that perfectly encapsulates the exact set of concepts you're asking about)

But LLMs are not reliable enough, so you can not actually expect “specificity”

Not perfect now, but adequate in some domains. Will only get better.

> Will only get better.

AI companies are still in their "burning money" phase.

Enshittification is not on the horizon yet, but it's inevitable.

While I have no doubt that individual companies, such as OpenAI for example, will eventually introduce enshittification features, I doubt the industry as a whole can be summarized that easily.

I believe, over all, development will go forward and things will get better. A rising tide lifts all ships, even if some of them decide to be shitty leaking vessels. If nothing else we always have open source software to fall back on when the enshittification of the proprietary models start.

For a practical example: The cars we drive today are a lot better than 100 years ago. A bad future isn't always inevitable.

But are cars we drive today better than they were 10 years ago? 20 years ago? Reliability is now trending down for cars, safety is questionable as deaths and injuries have been steadily increasing for a number of years now. Features are getting converted to subscriptions, and they all constantly send back telemetry.

Good points!

More reliable than 80% of my teachers growing up.

That is true in a broader sense, but education and abundant money don't generally go hand in hand.

don't I know it

"Having trouble with your algebra? MathWiz is having a 20% discount this month only. Only $24.95 / month. This is an excellent deal. Don't you want to improve? Do you want to let your family down, like they thought you would? Or would like me to create an account for you?"

"Want to get a job? [COLLEGE] is having a 0% discount -- only $200,000 a year! Don't you want to have a place to live? Go horribly in debt, pick a degree that may not matter, all at an age where your Brain has not yet developed fully!"

etc

> where people need to generate hundreds or thousands of these images

Anyone using image gen for real work not just for fun.

Although you're way better off finding your own workflows with local models at that scale.

Interior design, fashion, and advertising all come to mind

Imagine a news feed that never ends full of AI slop to sell ads on

Imagine an AI recipe building app that helps you create a recipe with certain ingredients, then generates an image of what the final product might look like.

Why do need to know what it looks like? Or are you publishing the recipe without cooking it?

Reply on news.ycombinator.com