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

Operator research preview

openai.com

Post reply on HN
Print view View original

441–448 of 448 posts

Re: Operator research preview

#441

Post by BrenBarn »

dang wrote: post #421
edouard-harris wrote: post #366

Earlier quoted context omitted.

By definition, in that limit they'll be genuinely adding to the discourse so presumably they should stay. Edit: More correctly, they'll be making contributions to the discourse that closely mimic the human distribution, so from a pure content perspective they won't be making the discourse any worse in the very short term.

I made a similar point a while ago (maybe last year) and there were some pretty good objections to it. Unfortunately I couldn't find that post when I looked for it last night!

One obvious counterpoint is that using AI tools allows manipulation of the discussion in a similar way to using a bullhorn in a coffee shop, only without revealing that you're the one holding the bullhorn. 10,000 bots "contributing to the discourse" in accordance with prompts and parameters controlled by one or a few individuals is quite different from the genuine opinions of 10,000 actual humans, especially if anyone is trying to use the discussion to get a sense of what humans actually think.

Re: Operator research preview

#442

Post by dang »

BrenBarn wrote: post #441
dang wrote: post #421

Earlier quoted context omitted.

I made a similar point a while ago (maybe last year) and there were some pretty good objections to it. Unfortunately I couldn't find that post when I looked for it last night!

One obvious counterpoint is that using AI tools allows manipulation of the discussion in a similar way to using a bullhorn in a coffee shop, only without revealing that you're the one holding the bullhorn. 10,000 bots "contributing to the discourse" in accordance with prompts and parameters controlled by one or a few individuals is quite different from the genuine opinions of 10,000 actual humans, especially if anyon…

That's a good counterpoint and, IIRC, it's in addition to the other good counterpoints that I still can't find.

Re: Operator research preview

#443

Post by itskarad »

null0pointer wrote: post #312
itskarad wrote: post #164

I think this opens a new direction in terms of UI for companies like Instacart or Doordash — they can now optimise marketing for LLMs in place of humans, so they can just give benchmarks or quantized results for a product so the LLM can make a decision, instead of presenting the highest converting products first. If the operator is told to find the most nutritious eggs for weight gain, the agent can refer to the nutr…

No problem, itskarad. I’ve ordered a pallet of Cadbury Creme Eggs to your house due for delivery tomorrow morning.

hahaha love it

Re: Operator research preview

#445

Post by rhubarbtree »

refulgentis wrote: post #396
rhubarbtree wrote: post #337

Earlier quoted context omitted.

What if the agent runs remotely?

Zugzwang - now I either need the user to preload all possible info/credentials and persistent containers if I want them to avoid having to do it again, or if I want to avoid paying some startup costs of ex. initializing git repos. Which is totally possible! Just...might as well do CLI first.

Thinking about "Mighty"... if their timing had been better...

Re: Operator research preview

#446

Post by jonsantillan »

Sometimes it feels like YC is the factory where OpenAI sources its ideas. Most features OpenAI launches seem to echo something a YC company built first. Inspiration, influence, or something deeper? After all, doesn’t YC have a stake in OpenAI?

Re: Operator research preview

#447

Post by raymond_goo »

swatcoder wrote: post #265
Workaccount2 wrote: post #261

Earlier quoted context omitted.

In swatcoders defence, he likely loaded the comments between the time that karpathy posted the AI comment and the time he posted the follow-up (there is a 5 minute gap). The timings on the comments seem to suggest that possibility at least.

Excellent forensics! You are correct. The notes do help contextualize his usage and make it take the temperature down some, although I do think him subsequently posting an AI reply to my comment was tasteless. (But I also get it. I used harsh words there and invited some ribbing in return.)

I thought it was hilarious! :-)

Re: Operator research preview

#448

Post by siliconwrath »

rad_gruchalski wrote: post #428
siliconwrath wrote: post #316

Earlier quoted context omitted.

With a true GPT ordering experience, you would just say “a Big Mac and a diet coke please” to a speaker just like you would in a drive thru and it would ring you up. It would replace the cashier

The guy taking orders does other things rather than just taking an order. Wake me up when chatgimp can prepare my fries and bring the bag with ready food to my car.

If the guy taking orders doesn’t have to take orders in addition to his other work anymore, they can hire less staff overall.
Post reply on HN