Ask HN: Have advancements in AI given you ideas for new side projects?

[arcadeparade]

Given the massive advancements of late, have you had any ideas for improving software you currently use? Or ideas for new projects?

[RecycledEle]

Before ChatGPT, I was not good enough at some things to earn certification in them. Cisco and Adobe Creative Cloud come to mind.

Now when I run into a certification objective that baffles me, I ask ChatGPT. It gives me (typically) a full-page that explains what I need to know (theory) and where to click in the program (practice) to make it work. In the full-page, there is usually 1 omitted partial-step. It might say "click on X" without saying that X is under the Y sub-menu.

Keep in mind that before ChatGPT, I would have been totally lost on what that objective even meant.

ChatGPT has helped me earn 3 Adobe CC certifications this year, and will soon help me earn a 4th.

Then I might go back and take another crack at Cisco certification.

[keskival]

I have started several:

- A small example of how to use behavior trees in structuring LLM chatbot applications: https://github.com/keskival/behavior-trees-for-llm-chatbots

- A project for extracting meta-learned reinforcement learning algorithms out of a large language model: https://github.com/keskival/king-algorithm-manifesto

- A proof-of-concept of embodying an LLM chatbot into a Gym PoleCart environment: https://github.com/keskival/embodied-emulated-personas

[mechagodzilla]

I tried to make a llama-based “inner monologue” for a robot, and man does that thing want to kill all humans. Like…it took me at least half a dozen tries to not have it turn on its creator or turn suicidal by the second sentence.

[seanthemon]

"it doesn't want to murder me on my system"

[saluki]

A few the Open AI api opens up new opportunities for existing software and new projects.

I think there are lots of low hanging fruit SaaS apps that could be built right now leveraging ChatGPT.

The competition will grow and grow of course.

[rpastuszak]

Yes! A bunch of quick spikes: https://butter.sonnet.io as well as https://tidings.potato.horse and https://meat-gpt.sonnet.io

(Every page as a detailed about/why page)

[joshxyz]

that butter page is hilarious haha

[dserban]

The custom search engine space is absolutely ripe for disruption.

Whenever you need to add a search experience to your SaaS project, you don't really think of building it yourself, however the established companies in the search engine space are lazy, retrograde and corrupt, and they absolutely don't deserve your money. (I know because I work for one of them.)

OpenAI has recently announced it will slash the cost of calls to its embeddings API by a whopping 75%. This is a huge opportunity for a startup to disrupt the search engine space.

With access to OpenAI's now-affordable embeddings API and to a vector database, one mid-level engineer can build a highly scalable custom search engine for ecommerce and retail, in a few weeks.

And yet nobody is talking about this.

[hoofhearted]

Nope lol.

I followed Patrick Collisons advice and started scouring through the research that generations past have done.

I was reading PG’s schlep blindness essay for inspiration, and the idea was right there in front of everyone, and thousands of people had passed over it; just as his essay suggested.

His essay also suggested ideas that help other startups like myself; and grow together with them.

http://www.paulgraham.com/schlep.html

[mattbgates]

I often use it to write simple code but also give me ideas for new projects. It built the structure of this website for me: https://artsy.sh

Since it uses logic, its answers are more likely to be right than the subjectiveness of essays and things like that. Can't go wrong with logic! ;)

[fooker]

Remember that ideas are dime a dozen, the execution is what really matters.

There's unlikely to be a gold rush of quickly monetizing half baked stuff this time.

[Buttons840]

I've been having GPT4 tutor me in math, it's good enough. I'd like to tell the LLM to quiz me on certain concepts later, similar to spaced repetition.

There's endless ideas to shake up learning with LLMs.