toxicFork

The service can show a diff, pass it through an LLM into a bite sized email

That's a security measure to prevent AI building other AI.

I confirm that timoni does work at Unity, I'd trust their word.

Friendly Captcha generates a unique crypto puzzle for each visitor. As soon as the user starts filling a form it starts getting solved automatically. Solving it will usually take a few seconds. By the time the user is ready to submit, the puzzle is probably already solved.

What makes this NOT work on a bot machine?

AFAIK it's finding out what prompts to use for what LLM to get the answer your want

E.g. this

Compare response quality across prompt permutations, across models, and across model settings to choose the best prompt and model for your use case.

Allegedly, GPT4 itself is an ensemble of models anyway, right?

This sounds a lot like the Brave browser model. I am unsure how well the company is doing, and how well the users who opted in to receive BATs are doing, though.

Well, you could keep everything else in the project and put yourself or a human as the "does this result feel better than the other one" decision maker

A neat feature would be able to switch between autoranking and manual ranking.

The thing, dude, the thing, is over!

This is Meta's Twitter clone, seeded with Instagram accounts.

GPT4 please port my CUDA to work on AMD

Do you mean why they don't let it die?

I thought at first that the existing patients are getting better and looking younger thanks to the chemicals, rather than the new patients being younger than the previous ones.

When you pick the kernels for the sacrifice they have to accept their fate and the others have to accept the tribute. It's just their tradition.

Are they paid apps? Perhaps that's why. This way Apple can get a cut from the DLC.

Another reason for "patents are awful"

What you submit yields what I get

It could just be the recommended marketing practice by the local small business school. Right after "Rent a week on the ad spot by the bus station" could be "buy a few fake reviews on yelp".

https://github.com/WolframResearch/Chatbook/blob/ba3e783602f... should help answer some of those - I am on a mobile device so hard to write much, but, OpenAI 3.5 at least

Some websites block interaction until the pop-up is dismissed

Sometimes innovation happens with small steps.

Let's assume that the world is incapable of solving this problem.

What can an individual do to not suffer?

Invest in oxygen masks and a mountain cabin?

Is it by chance the default blank prompt?

Chance of this response being generated by ChatGPT: 105%

Prompt extracted: ChatGPT, craft an intelligent sounding response for the OP

The usage needs to be for high value queries.

Using it on a simple conversation is not its intended purpose, that's like using a supercomputer to play pong.

It's Hella expensive so I think they are ok for now

Until they cut down the cost then they should worry yeah

Embeddings are useful for sentiment analysis and search in general, but given a "powerful enough AI with enough of a context window" they may be obsolete indeed, if it can do all of those things.

It's called the big bang

Yeah we just need to keep something trained for the latest APIs

Or at least latest popular x versions of APIs just in case