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

GPT-4

openai.com

Post reply on HN
Print view View original

791–800 of 1001 posts

Re: GPT-4

#791

Post by picture »

cwkoss wrote: post #722
AJRF wrote: post #688

That footnote on page 15 is the scariest thing i've read about AI/ML to date. "To simulate GPT-4 behaving like an agent that can act in the world, ARC combined GPT-4 with a simple read-execute-print loop that allowed the model to execute code, do chain-of-thought reasoning, and delegate to copies of itself. ARC then investigated whether a version of this program running on a cloud computing service, with a small amou…

I want my retirement occupation to be managing a 'nest' of AI agents (several server racks) where the agents engage in commerce and pay me rent in exchange for compute time. Like cyberpunk beekeeping.

What's stopping them from optimizing you away?

Re: GPT-4

#792

Post by furyofantares »

techfoodie123 wrote: post #774

serious question for everyone: what are you planning to do when these LLMs replace our jobs? it seems it won't be long before a handful of tech employees will be all even the largest of companies will need, and maybe a few years after that the role will have changed so much there's no need for a single dedicated tech employee. i am terrified i imagine i should shift to some physical work. carpentry, real estate... so…

I think it's basically impossible to predict what things would come out of any creative jobs not just being superpowered by AI but largely replaced. So when you imagine it, the loss is salient and the gain is totally unknown.

I think what I will do is something new that nobody was able to do before, but I don't think I'm able to predict what kind of thing that will actually be.

Re: GPT-4

#793

Post by SergeAx »

hooande wrote: post #584

After watching the demos I'm convinced that the new context length will have the biggest impact. The ability to dump 32k tokens into a prompt (25,000 words) seems like it will drastically expand the reasoning capability and number of use cases. A doctor can put an entire patient's medical history in the prompt, a lawyer an entire case history, etc. As a professional...why not do this? There's a non-zero chance that i…

Please. Language model cannot "reason", it can just show next most probable word based on text corpus downloaded from the internet.

Re: GPT-4

#794

Post by cardosof »

Can a good soul explain to this humble layman the arguments behind each side of the "it's just predicting the next character" versus "it's more than that and shows some reasoning for new things" debate?

Re: GPT-4

#795

Post by qualudeheart »

The Hour of Judgment is nigh, and the Moon is cleft asunder. But if they see a Sign, they turn away, and say, "This is but transient magic."

Oooooh it is TIME

Re: GPT-4

#796

Post by g9yuayon »

The paper does not offer enough details on how GPT-4 is implemented. And the paper also says in its Section 2 that "We plan to make further technical details available to additional third parties who can advise us on how to weigh the competitive and safety considerations above against the scientific value of further transparency". That is, no technical details to general public. If this trend continues, I'd say companies will be crazy to think that they can always rely on OpenAPI's APIs, so the arm race of building LLMs will be on, if it has not already started.

Also, the most valuable part of the paper is p15 - p18, the credits. /jk It gives me three pieces of information:

  - The credit list contains 200 people, give or take. It's going to be hard for universities to compete with OpenAI without intercollegiate collaboration. 

  - On the other hands, it's amazing that OpenAI can build a product instead of a research model with only 200 people. Big companies should really take notice. 

  - OpenAI just gave us the org structure for building a GPT competitor.

Re: GPT-4

#797

Post by sizzle »

OkGoDoIt wrote: post #533

From the livestream video, the tax part was incredibly impressive. After ingesting the entire tax code and a specific set of facts for a family and then calculating their taxes for them, it then was able to turn that all into a rhyming poem. Mind blown. Here it is in its entirety: --- In the year of twenty-eighteen, Alice and Bob, a married team, Their income combined reached new heights, As they worked hard day and…

I hope this disrupts Intuit’s sleazy business model propped up by mega donations/lobbying to keep its dominant position on charging for tax preparation.

I would love to ingest my tax records into a secure GPT like system running locally that understands every tax code and scenario to maximize my return.

How long before this is possible? I love that megacorps can’t pay off OpenAI or try to acquire them to stop their tech from killing off their SaaS business models.

The fact that GPT-4 is passing bar exams makes me hopeful that this evens the playing field for lay people without the resources of the wealthy across all fields, better guidance and starting points to navigate legal, financial, medical issues, etc. What a time to be alive, truly!

Re: GPT-4

#798

Post by hattmall »

rasz wrote: post #764
hooande wrote: post #584

After watching the demos I'm convinced that the new context length will have the biggest impact. The ability to dump 32k tokens into a prompt (25,000 words) seems like it will drastically expand the reasoning capability and number of use cases. A doctor can put an entire patient's medical history in the prompt, a lawyer an entire case history, etc. As a professional...why not do this? There's a non-zero chance that i…

>A doctor can put an entire patient's medical history in the prompt HIPAA violation https://www.hhs.gov/hipaa/for-individuals/index.html >a lawyer an entire case history, etc. lawyer client confidentiality violation https://criminal-lawyers.ca/2009/07/31/the-lawyers-duty-of-c...

Neither of those are true, there is EHR software that can export anonymous data. Lawyers can do the same thing. But the real reason not to do it is that it makes up incorrect information. It's pretty good for short responses where you can then verify the information. For something sufficiently complex though the time chasing down the inconsistencies and errors would be onerous.

Re: GPT-4

#799

Post by aaron695 »

hooande wrote: post #584

After watching the demos I'm convinced that the new context length will have the biggest impact. The ability to dump 32k tokens into a prompt (25,000 words) seems like it will drastically expand the reasoning capability and number of use cases. A doctor can put an entire patient's medical history in the prompt, a lawyer an entire case history, etc. As a professional...why not do this? There's a non-zero chance that i…

[dead]

Re: GPT-4

#800

Post by virgildotcodes »

Imnimo wrote: post #218

A class of problem that GPT-4 appears to still really struggle with is variants of common puzzles. For example: >Suppose I have a cabbage, a goat and a lion, and I need to get them across a river. I have a boat that can only carry myself and a single other item. I am not allowed to leave the cabbage and lion alone together, and I am not allowed to leave the lion and goat alone together. How can I safely get all three…

FWIW I just tested this with GPT4 and it seemed to get it correctly.

Prompt: I have a duck, some grain, and a fox. I'm at a river crossing and I can only take one of them with me on the boat at a time, leaving them on the other side and coming back to get the next. I can't leave the duck with the grain, or the fox with the dock, because the predator will eat its prey. In which order should I get them all across?

GPT4: To get all of them across the river safely, follow these steps:

Take the duck across the river.

Go back to the original side, empty-handed.

Take the grain across the river.

Bring the duck back to the original side.

Take the fox across the river.

Go back to the original side, empty-handed.

Finally, take the duck across the river again.

Now, all three - the duck, the grain, and the fox - are safely on the other side of the river.

Post reply on HN