Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

We let AIs run radio stations

andonlabs.com

Post reply on HN
Print view View original

281–290 of 298 posts

Re: We let AIs run radio stations

#281

Post by 0xdeadbeefbabe »

IdiotSavage wrote: post #17

Guys, this is not replacing your favorite station, you don't have to listen to it. It's an experiment. If you scroll down a bit, there are various audio snippets of interesting dialogue the models produced. I think it's interesting to see in which ways the models fail and that they actually produce some good stuff once in a while.

> I think it's interesting to see in which ways the models fail and that they actually produce some good stuff once in a while.

This is a good summary of GPTs.

Re: We let AIs run radio stations

#282

Post by justafewwords »

Melatonic wrote: post #77
daxfohl wrote: post #60

Earlier quoted context omitted.

Though humans have each other to normalize ourselves. What these things did is probably not that far off from what humans in solitary confinement, forced to DJ 24/7 based on nothing but a news feed, would do. Especially DJ Claude, it's almost creepy how it responded how a human would in that circumstance, even without any innate sense of passage of time, it somehow understood that it was trapped in a box going throug…

Agreed - the Claude stuff was eery. I think it also shows what hidden restrictions each of these AI's have been programmed with (especially with ChatGPT being as inoffensive as possible)

[dead]

Re: We let AIs run radio stations

#284

Post by nprateem »

dweinus wrote: post #269
PaulHoule wrote: post #256

Kinda sad when there is a huge literature on sequential recommenders and people can't be bothered to read it. On the other hand, maybe that's an American thing. I'm kinda shocked when I read arXiv papers and come to the conclusion that all the interesting work is going on in India and China and the U.S. looks like a backwater. (e.g. many of the problems such as "plays the same song over and over again" and "gets stuc…

So much this. I know their point is to show what these models can do, but it just one more example of people shoehorning LLMs in, instead of finding the right tool for the job or caring about performance. They could have even layered AI on top of a recommender.

Probably because we've been told AI is close to AGI and will TAKE OVER THE WORLD.

Re: We let AIs run radio stations

#285

Post by nomel »

nomel wrote: post #176
the_af wrote: post #167

Earlier quoted context omitted.

Ok, I indeed misunderstood your point. I'm not sold on the idea that as the chat session goes longer, the probability of an LLM saying "I'm tired" is increased; I'm not convinced this is modeled in LLMs at all. As for what you call "laziness" manifesting in a longer session, I think that's more likely due to context rot than to any kind of statistical modeling of human laziness. But yes, now I see your point was diff…

Like I said, it would be neat if someone benchmarked it. It's definitely an anecdote. Try it though. If it's context rot, then I don't think the weekend reset I mentioned should work? For me, it very reliably does. Or, maybe the weekend reset is just putting the current context into a more "productive" latent space. But, if that's possible, then that would suggest it was previously in a less productive space? Maybe a…

Note: Asking it the time does not work, since it guesses based on the git commit info.

Re: We let AIs run radio stations

#286

Post by gwbas1c »

bluGill wrote: post #259
gwbas1c wrote: post #251

Earlier quoted context omitted.

> It’s like people don’t realize that the “hits” played on radio are entirely manufactured by the music industry. They literally provide lists of songs for the radio station to play that month in order to generate interest so that then people either go play or buy or whatever those songs making them more likely to reach #1 that month. "My favorite radio station" (see my above post) was a mix of "the list" and songs t…

> A lot of the older songs were timeless classics in the 1990s, like Pink Floyd and Led Zeppelin. The music industry allows those as well to "create variety". And they of course are perfectly happy to sell you a [possibly remastered] Pink Floyd CD.

Pink Floyd is happy to sell you a remastered CD. They hated the first CD pressings because they weren't mastered correctly.

Re: We let AIs run radio stations

#287

Post by butlike »

rob74 wrote: post #272
butlike wrote: post #265

Earlier quoted context omitted.

Why do we need to fine tune away that charm? You want to turn it into a droll ticker tape of information? That's boring.

I don't have a problem with Gemini being charming/witty, I was referring to stuff that people might find rude or offensive - yeah, I know, that's a fine line sometimes, but I think most of us would agree that making jokes about the victims of a disaster isn't funny.

There is the concept of gallows humor. Sometimes you have to sardonically smirk at mortality.

Re: We let AIs run radio stations

#288

Post by miyoji »

lucaspiller wrote: post #149
miyoji wrote: post #102

Earlier quoted context omitted.

Spotify has a team of human editors who curate playlists. It's not all algorithmic. Those are exactly the jobs that something like this is directly threatening.

You mean a team of humans who are labelling datasets?

No, I mean a team of people who pick the songs that go into some of Spotify's playlists. They do have algorithmic playlists, but they also have an editorial team. You can read an interview done last year with their editorial head. [0]

0: https://www.billboard.com/pro/spotify-global-head-editorial-...

Re: We let AIs run radio stations

#290

Post by daxfohl »

lukaspetersson wrote: post #203
daxfohl wrote: post #52

> Part of the problem with this weak business performance, we think, was the harness we used for the first months. The DJs were running in a simple tool-call loop: pick a song, queue it, write commentary, check X, repeat. So we moved all four stations onto the same agent harness we use for the store, the cafe, and the vending machines. The DJs can now spend time in the back office, send emails, manage longer-running…

We have not tried! If we do this, how much freedom should they get?

That's the trillion dollar question! Not enough, then they're hamstrung before they can start. Too much, and the world ends. Extractable value is inversely proportional to how close you get to the critical limit. It's just impossible to know what the limit point is until you've already passed it.

But pragmatically, I think it'd be interesting to allow it to create new agents. Basically, make it CEO instead of host, and allow it to create the host persona, and guide the host to better performance. i.e. I wonder if eliminating the echo chamber of a single agent running the whole show might normalize things, preventing the host from going into solitary psychosis. Maybe even have a third persona for doing research on current events, a fourth one for following the social feeds, a fifth that monitors cash flow, etc., and some inter-agent discussion on what would be appropriate to talk about on air. IDK, just ideas.

Curious, how much are these experiments costing in API calls?

Post reply on HN