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

GPT-5 is behind schedule

wsj.com

Post reply on HN
Print view View original

521–530 of 1001 posts

Re: GPT-5 is behind schedule

#521

Post by zeroonetwothree »

williamcotton wrote: post #78
mathieuh wrote: post #26

Earlier quoted context omitted.

It’s the same for me. I genuinely don’t understand how I can be having such a completely different experience from the people who rave about ChatGPT. Every time I’ve tried it’s been useless. How can some people think it’s amazing and has completely changed how they work, while for me it makes mistakes that a static analyser would catch? It’s not like I’m doing anything remarkable, for the past couple of months I’ve b…

I found it very useful for writing a lexer and parser for a search DSL and React component recently: https://github.com/williamcotton/search-input-query

Interesting. I implemented something very similar (if not identical) a couple years ago (at work so not open source). I used a simple grammar and standard parser generator. It’s been nice to have the grammar as we’ve made tweaks over the years to change various behaviours and add features.

Re: GPT-5 is behind schedule

#522

Post by ramesh31 »

ferminaut wrote: post #440
IAmGraydon wrote: post #433

Earlier quoted context omitted.

Gemini 2.0 isn't particularly great at coding. The Gemini 1206 preview that was released just before 2.0 is quite good, though. Still, it hasn't taken the crown from Claude 3.5 Sonnet (which appears to now be tied with o1). Very much agree about Cline + VSCode, BTW. My preferred models with Cline are 3.5 Sonnet and 3.5 Haiku. I can throw the more complex problems at Sonnet and use Haiku for everything else. https://a…

the context limits on google are nuts! Being able to pump 2 million tokens in and having it cost $0 is pretty crazy rn. Cline makes it seamless to switch between APIs and isnt trying to shoehorn their SAAS AI into a custom vscode (looking at you cursor)

>the context limits on google are nuts! Being able to pump 2 million tokens in and having it cost $0 is pretty crazy rn.

What's the catch though? I was looking at Gemini recently and it seemed too good to be true.

Re: GPT-5 is behind schedule

#523

Post by BOOSTERHIDROGEN »

SamPatt wrote: post #415

I'm sure the debate over the definition of AGI is important and will continue for a while, but... I can't care about it anymore. Between Perplexity searching and summarizing, Claude explaining, and qwen (and other tools) coding, I'm already as happy as can be with whatever you want to call this level of intelligence. Just today I used a completely local AI research tool, based on Ollama. It worked great. Maybe it won…

Can you walk me through the steps you've taken to set up the Ollama-based tool so far?

Re: GPT-5 is behind schedule

#524

Post by wruza »

eru wrote: post #417
worik wrote: post #387

Earlier quoted context omitted.

If that is what AGI looks like. There may well be an upper limit on cognition (we are not really sure what cognition is - even as we do it) and it may be that human minds are close to it.

Yes, we can imagine that there's an upper limit to how smart a single system can be. Even suppose that this limit is pretty close to what humans can achieve. But: you can still run more of these systems in parallel, and you can still try to increase processing speeds. Signals in the human brain travel, at best, roughly at the speed of sound. Electronic signals in computers play in the same league as the speed of ligh…

I think our issue is much more banal: we are very slow talkers and our effective communication bandwidth is measured in bauds. Anything that could bridge this airgap would fucking explode in intelligence.

Re: GPT-5 is behind schedule

#525

Post by acchow »

SamPatt wrote: post #415

I'm sure the debate over the definition of AGI is important and will continue for a while, but... I can't care about it anymore. Between Perplexity searching and summarizing, Claude explaining, and qwen (and other tools) coding, I'm already as happy as can be with whatever you want to call this level of intelligence. Just today I used a completely local AI research tool, based on Ollama. It worked great. Maybe it won…

Cline was fixing my type errors and unit tests while I was doing my V60 pourover.

Re: GPT-5 is behind schedule

#526

Post by nico »

SamPatt wrote: post #415

I'm sure the debate over the definition of AGI is important and will continue for a while, but... I can't care about it anymore. Between Perplexity searching and summarizing, Claude explaining, and qwen (and other tools) coding, I'm already as happy as can be with whatever you want to call this level of intelligence. Just today I used a completely local AI research tool, based on Ollama. It worked great. Maybe it won…

> Just today I used a completely local AI research tool, based on Ollama. It worked great

What’s it called? Could you post a link please?

Thank you

Re: GPT-5 is behind schedule

#528

Post by Jean-Papoulos »

Animats wrote: post #123

"Orion’s problems signaled to some at OpenAI that the more-is-more strategy, which had driven much of its earlier success, was running out of steam." So LLMs finally hit the wall. For a long time, more data, bigger models, and more compute to drive them worked. But that's apparently not enough any more. Now someone has to have a new idea. There's plenty of money available if someone has one. The current level of LLM…

> So LLMs finally hit the wall. For a long time, more data, bigger models, and more compute to drive them worked

We can't say whether there is a wall, since we don't have anymore data to train on.

Re: GPT-5 is behind schedule

#529

Post by vlovich123 »

zeroonetwothree wrote: post #515
stravant wrote: post #463

Earlier quoted context omitted.

I disagree because AI only has to get good enough at doing a single thing: AI research. From there things will probably go very fast. Self driving cars can't design themselves, once AI gets good enough it can

It’s possible (maybe even likely) that “AI research” is “AGI-hard” in that any intelligence that can do it is already an AGI.

It’ll probably sit in the human hybrid phase for longer than with chess where the AGI tools make the humans better and faster. But as long as the tools keep getting better at that there’s a strong flywheel effect

Re: GPT-5 is behind schedule

#530

Post by TeMPOraL »

sollewitt wrote: post #493
atonse wrote: post #457

Earlier quoted context omitted.

Same here. The ability to “talk to an expert” about any topic I’m curious about and ask very specific questions has been invaluable to me. It reminds me of being a kid and asking my grandpa a million questions, like how light bulbs worked, or what was inside his radio, or how do we have day and night. And before anyone talks about accuracy or hallucinations, these conversations usually are treated as starting off poi…

LLMs suffer from the "Igon Value Problem" https://rationalwiki.org/wiki/Igon_Value_Problem Similar to reading a pop sci book, you're getting an entertainment from a thing with no actual understanding of the source material rather than an education.

Oh so you mean I have at my fingertips a tool that can generate me a Scientific American issue on any topic I fancy? That's still some non-negative utility right there :).
Post reply on HN