19 ms·

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 a

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't get much better? Or maybe it'll take decades instead of years? Ok. I remember not having these tools. I never want to go back.

nico 2y ago
> 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
- SamPatt 2y ago
  Here you go https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ollama https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ol...
acchow 2y ago
Cline was fixing my type errors and unit tests while I was doing my V60 pourover.
ferminaut 2y ago
vscode + cline extension + gemini2.0 is pretty awesome. Highly recommend checking out cline. it quickly became one of my favorite coding tools.
- SamPatt 2y ago
  I will check it out. The number of new tools is staggering. I enjoy image and video generation and I have a 4090 and ComfyUI; I can't keep up with everything coming out anymore.
  - ferminaut 2y ago
    If you're a offline kind of guy, try LM Studio + Cline :) /not affiliated with cline, just a happy user
  - IAmGraydon 2y ago
    If you're interested in the latest tools for coding, join this subreddit and you'll always be on top of it: https://www.reddit.com/r/ChatGPTCoding/ https://www.reddit.com/r/ChatGPTCoding/ There are a lot of tools, but only a small pool of tools that are worth checking out. Cline, Continue, Windsurf, CoPilot, Cursor, and Aider are the ones that come to mind.
    - deadmutex 2y ago
      "ChatGPT" Coding... is it impartial? the name sorta sounds biased.
      - IAmGraydon 2y ago
        ChatGPT was the first to come along, so the subreddit was given a perhaps short-sighted name. It's now about coding with LLMs in general.
- IAmGraydon 2y ago
  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://aider.chat/docs/leaderboards/edit.html https://aider.chat/docs/leaderboards/edit.html
  - ferminaut 2y ago
    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)
    - ramesh31 2y ago
      >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.
      - Jensson 2y ago
        Google inference is a lot cheaper since they have their own hardware so they don't have to pay licensing to NVIDIA, thus their free tier can give you much more than others. Other than that the catch is like all other free tiers, it is marketing and can be withdrawn at any moment to get you to pay after you are used to their product.
      - HyprMusic 2y ago
        Your code becomes training data[0]: > When you use Unpaid Services, including, for example, Google AI Studio and the unpaid quota on Gemini API, Google uses the content you submit to the Services and any generated responses to provide, improve, and develop Google products and services and machine learning technologies, including Google's enterprise features, products, and services, consistent with our Privacy Policy. [0] https://ai.google.dev/gemini-api/terms https://ai.google.dev/gemini-api/terms
IAmGraydon 2y ago
I've bee thinking the same thing lately. Even if we don't get to AGI, LLMs have revolutionized the way I work. I can produce code and copy at superhuman speeds now. I love it. Honestly, if we never get to AGI and just have the LLMs, it's probably the best possible outcome as I don't think true AGI is going to be a good thing for humanity.
lazygoose 2y ago
Curious about the AI research tool you mentioned, would you mind sharing it? Been trying to get a good local research setup with Ollama but still figuring out what works best.
- SamPatt 2y ago
  https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ollama https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ol...
- Bilal_io 2y ago
  Not OP, but based on their mention of Ollama, I can tell you that it has built in search tools, all you need to do is supply an API to one of the tools, or even run one of the search tools locally using docker.
divan 2y ago
Let's revisit this comment in one year – after the explosion of agentic systems. (:
- wokwokwok 2y ago
  We already have agentic systems; they're not particularly impressive (1). There's no specific reason to expect them to get better. Things that will shift the status quo are: MCST-LLMs (like with ARC-AGI) and Much Bigger LLMs (like GPT-5, if they ever turn up) or some completely novel architecture. [1] - It's provable; if just chaining LLMs are a particular size into agentic systems could scale indefinitely, then you could use a 1-param LLM and get AGI. You can't. QED. Chaining LLMs with agentic systems has a capped maximum level of function which we basically already see with the current LLMs. ie. Adding 'agentic' to your system has a finite, probably already reached, upper bound of value.
  - NitpickLawyer 2y ago
    > It's provable; if just chaining LLMs are a particular size into agentic systems could scale indefinitely, then you could use a 1-param LLM and get AGI. You can't. QED. Perhaps I missunderstand your reply, but that has not been my experience at all. There are 3 types of "agentic" behaviour that has worked for a while for me, and I don't know how else it would work without "agents": 1. Task decomposition - this was my manual flow since pre-chatgpt models: a) provide an overview of topic x with chapter names; b) expand on chapter 1 ... n ; c) make a summary of each chapter; d) make an introduction based on the summaries. I now have an "agent" that does that w/ minimal scripting and no "libraries". Just pure python control loop. This gets me pretty reasonable documents for my daily needs. 2. tool use (search, db queries, API hits). I don't know how you'd use an LLM without this functionality. And chaining them into flows absolutely works. 3. coding. I use the following "flow" -> input a paragraph or 2 about what I want, send that + some embedding-based context from the codebase to an LLM (3.5 or 4o, recently o1 or gemini) -> get code -> run code -> /terminal if error -> paste results -> re-iterate if needed. This flow really works today, especially with 3.5. In my testing it needs somewhere under 3 "iterations" to "get" what's needed in more than 80% of the cases. I intervene in the rest of 20%.
atonse 2y ago
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 points to then start googling specific terms, people, laws, treaties, etc to dig deeper and verify. Last year during a visit to my first Indian reservation, I had a whole bunch of questions that nobody in person had answers to. And ChatGPT was invaluable in understanding concepts like where a reservation’s autonomy begins and ends. And why certain tribes are richer than others. What happens when someone calls 911 on a reservation. Or speeds. Or wants to start a factory without worrying about import/export rules. And what causes some tribes to lose their language faster than others. And 20 other questions like this. And most of those resulted in google searches to verify the information. But I literally could never do this before. Same this year when I’m visiting family in India. To learn about the politics, the major players, WHY they are considered major players (like the Chief Minister of Bengal or Uttar Pradesh or Maharashtra being major players because of their populations and economies). Criticisms, explanations of laws, etc etc. For insanely curious people who often feel unsatisfied with the answers given by those around them, it’s the greatest thing ever.
- stravant 2y ago
  One of my favorite successes was getting an LLM to write me a program to graph how I subjectively feel the heat of steam coming off of the noodles I'm pouring the water out from as a function of the ambient temperature. I was wondering which effects were at play and the graph matched my subjective experience well.
- 1209412comb 2y ago
  Semi-related but I find that sometime it just completely ruined a type of conversation. Like as in your example, I would previously asked people "how would 911 handle an US Reservation Area", and watch how my friends think and reason. To me getting a conclusive answer was not a point. Now they just copy & paste Chat GPT, no fun haha.
bloppe 2y ago
At this point I think even the most bearish have to concede that LLM's are an amazing tool. But OpenAI was never supposed to be about creating tools. They're supposed to create something that can completely take over entire projects for you, not just something that can help you work on the projects faster. If they can't pull that off in the next year or two, they're gonna seriously struggle to raise the next 10B they'll need to keep the lights on. Of course LLMs aren't going anywhere, but I do not envy Sam Altman right now.
- lumost 2y ago
  At this point it’s quite likely that they could pivot and just be the chatgpt company. I’ve found chatgpt-4o with web search and plugins to be more useful than o1 for most tasks. It’s possible we’re nearing the end of the LLM race, but I doubt that’s the end of the AI story this decade, or OpenAI.
  - bloppe 2y ago
    Ya I think they probably will, but "the chatgpt company" is not worth 157B. It might not even be worth 1B.
    - josu 2y ago
      It has replaced ~50% of my Google searches.
      - silisili 2y ago
        Yes but it also hasn't been attacked by ads yet. Google doesn't suck for lack of search results, it sucks because of ads. Imagine asking chatgpt to tell you about slopes in Colorado, and the first five answers are about how awesome North Face is and how you can order from them. You probably wouldn't use it as much.
        jeffhuys 2y ago
        Local models are GOOD as well, and easy to use (ollama + open web ui). OpenAI has to perform a huge trick in order to stay relevant.
throwup238 2y ago
Amen. Everyone is talking about plateaus and diminishing returns on training but I don’t care one bit. I get that this is a startup focused forum and the financial sustainability of the market players is important but I can’t wait to see what the next decade of UX improvements will be like even if model improvements slow to a crawl.
uludag 2y ago
I'm pretty sure the plan has never been to just make these tools that make us more efficient. If AI stays at the level it's at, it would be a profound failure for companies like OpenAI. We're all benefiting from the capital being poured into these technologies now. The enshittification will come. The enshittification always comes.
- mitemte 2y ago
  I’m paying $240 a year to Anthropic that I wasn’t paying before and it’s worth it. While I don’t use Claude every single day, but I use it several times a day when I’m working. More times than the free tier allows.
  - aprilthird2021 2y ago
    Why do people say this like it's a refutation? Current valuation and investments were not based on getting a very small group of nerds (affectionately) on HN to pay $250/yr which probably doesn't cover even inference costs for the models let alone training and R&D
spaceman_2020 2y ago
This is me. If things never improve and Sonnet 3.6 is the best we have…I’m fine. Its good enough to drastically improve productivity
jayseattle 2y ago
[dead]
BOOSTERHIDROGEN 2y ago
Can you walk me through the steps you've taken to set up the Ollama-based tool so far?
alickz 2y ago
i feel like AGI is an arbitrary line in the sand anyway i think as humans we put too much emphasis on what intelligence means relative to ourselves, instead of relative to nature
rkagerer 2y ago
completely local AI research tool, based on Ollama Could you elaborate? Was it easy to install?
- arcanemachiner 2y ago
  Not OP, but yeah, ollama is super easy to install. I just installed the Docker version and created a little wrapper script which starts and stops the container. Installing different models is trivial. I think I already had CUDA set up, not sure if that made a difference. But it's quick and easy. Set it up, fuck around for an hour or so while you get things working, then you've got your own local LLM you can spin up whenever you want.
  - foobiekr 2y ago
    Does ollama still execute whatever arbitrary python code is in the model?
- SamPatt 2y ago
  Yes I was referring to this: https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ollama https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ol...
htrp 2y ago
> Just today I used a completely local AI research tool, based on Ollama. It worked great. Is it on github?
tugu77 2y ago
Thats all fine, but I think you are missing the bigger picture. It's not about whether what we already got out of this is good. Of course it is. But this is about where it's going. Until about 120 years ago, people were happy with horses and horse carriages. Such a great help! Travel long distances, pull weights, I never want to go back! But then the automobile was invented and within a few years little travel was done by horses anymore. More recently, everybody had a landline phone at home. Such great tech! Talk to grandma hundreds of miles away! I never want to go back! Then suddenly the mobile phone and just shortly after the smart phone came along and now nobody has a landline anymore but everybody can record tiktoks anywhere anytime and share them with the world within seconds. Now imagine "AI". Sure, we have some new tools right now. Sure we don't want to go back. But imagine the transformative effects that could come if the train didn't stop here. Question is just: will it?
hamilyon2 2y ago
If the progress in capabilities stall, the product fit, adoption, ease of use are the next battlefield. OpenAI may be first to realize and switch, so they still have a chance to recoup some of those billions
badgersnake 2y ago
They’re garbage, they will always be garbage. Changing a 4 to a 5 will not make it not garbage. The whole sector is a hype bubble artificially inflating stock prices.
- Elextric 2y ago
  https://www.youtube.com/watch?v=lFc1jxLHhyM https://www.youtube.com/watch?v=lFc1jxLHhyM
  - badgersnake 2y ago
    If that’s supposed to be impressive, it really isn’t.
ninetyninenine 2y ago
The definition of agi is a linguistic problem but people confuse it for a philosophical problem. Think about it. The term is basically just a classification and what features and qualities fit the classification is an arbitrary and linguistic choice. The debate stems from a delusion and failure to realize that people are simply picking and choosing different fringe features on what qualifies as agi. Additionally the term exists in a fuzzy state inside our minds as well. It’s not that the concept is profound. It’s that some of the features that define the classification of the term we aren’t sure about. But this doesn’t matter because we are basically just unsure about the definition of a term that we completely made up arbitrarily. For example the definition of consciousness seems like a profound debate but it’s not. The word consciousness is a human invention and the definition is vague because we choose the definition to be ill defined, vague and controversial. Much of the debate on this stuff is purely as I stated just a language issue.
- zifpanachr23 2y ago
  If it's genuinely what you say, then how is what is going on not slavery? I don't believe AGI is possible but if it was and it was as subjective as you say what is and isn't conscious, then it starts to take on an even more altogether evil character. Akin to cloning slave humans or something for free cheap labor.
  - ninetyninenine 2y ago
    how does a linguistic and language issue relate to slavery. It's the definition of a word. That's all. Slavery is also a word. Don’t you find it strange that your entire moral framework is constructed on top of the basis of arbitrary definitions of vocabulary? Make what you think is right or wrong based not off of language. Language is a delusion that masquerades as something with actual meaning when it is just an invention, a tool, to facilitate communication. Right now your concept of right and wrong is a vocabulary issue. Does this make sense? No.
delusional 2y ago
As a consumer you should always evaluate the product that is in front of you, not the one they promise in 6 months. If what's there is valuable to you, then that's great. When we discuss the potential AGI we're not talking as consumers, we're talking about the business side. If AGI is not reached, you'll see an absolutely enormous market correction, as it realizes that the product is not going to replace any human workers. The current generation of products are not profitable. They're investments towards that AGI dream. If that dream doesn't happen, then the current generation of stuff will disappear too, as it becomes impossible to provide at a cost you'd be comfortable with.
- rajamaka 2y ago
  Human workers workers have already been replaced.
ryukoposting 2y ago
At this point, most conceivable beneficial use cases for LLMs have been covered. If the economics of AI tech were aligned with making a good product that people want and/or need, we'd basically take everything we have at this point and make it lighter, smaller, and faster. I doubt that's what will happen.
michaelbuckbee 2y ago
What was the AI search tool?
tiffanyh 2y ago
I have the opposite reaction. AI right now feels like that MBA person at work. They don’t know anything. But because they sound like they are speaking with authority & confidence, allows them to get promoted at work. (While all of the experts at work roll their eyes because they know the MBA/AI is just spitting out nonsense & wish the company never had any MBA/AI people)
- zppln 2y ago
  And the MBA person (at my company this is everyone in middle management) is also the person who go around and suggest we shoehorn AI into everything...
fud101 2y ago
how do you interact with perplexity? mobile app?
ByteAndBattle 2y ago
Local search with Ollama? Please share!

nico 2y ago

> 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

SamPatt 2y ago
Here you go https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ollama https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ol...

acchow 2y ago

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

ferminaut 2y ago

vscode + cline extension + gemini2.0 is pretty awesome. Highly recommend checking out cline. it quickly became one of my favorite coding tools.

SamPatt 2y ago
I will check it out. The number of new tools is staggering. I enjoy image and video generation and I have a 4090 and ComfyUI; I can't keep up with everything coming out anymore.
- ferminaut 2y ago
  If you're a offline kind of guy, try LM Studio + Cline :) /not affiliated with cline, just a happy user
- IAmGraydon 2y ago
  If you're interested in the latest tools for coding, join this subreddit and you'll always be on top of it: https://www.reddit.com/r/ChatGPTCoding/ https://www.reddit.com/r/ChatGPTCoding/ There are a lot of tools, but only a small pool of tools that are worth checking out. Cline, Continue, Windsurf, CoPilot, Cursor, and Aider are the ones that come to mind.
  - deadmutex 2y ago
    "ChatGPT" Coding... is it impartial? the name sorta sounds biased.
    - IAmGraydon 2y ago
      ChatGPT was the first to come along, so the subreddit was given a perhaps short-sighted name. It's now about coding with LLMs in general.
IAmGraydon 2y ago
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://aider.chat/docs/leaderboards/edit.html https://aider.chat/docs/leaderboards/edit.html
- ferminaut 2y ago
  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)
  - ramesh31 2y ago
    >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.
    - Jensson 2y ago
      Google inference is a lot cheaper since they have their own hardware so they don't have to pay licensing to NVIDIA, thus their free tier can give you much more than others. Other than that the catch is like all other free tiers, it is marketing and can be withdrawn at any moment to get you to pay after you are used to their product.
    - HyprMusic 2y ago
      Your code becomes training data[0]: > When you use Unpaid Services, including, for example, Google AI Studio and the unpaid quota on Gemini API, Google uses the content you submit to the Services and any generated responses to provide, improve, and develop Google products and services and machine learning technologies, including Google's enterprise features, products, and services, consistent with our Privacy Policy. [0] https://ai.google.dev/gemini-api/terms https://ai.google.dev/gemini-api/terms

IAmGraydon 2y ago

I've bee thinking the same thing lately. Even if we don't get to AGI, LLMs have revolutionized the way I work. I can produce code and copy at superhuman speeds now. I love it. Honestly, if we never get to AGI and just have the LLMs, it's probably the best possible outcome as I don't think true AGI is going to be a good thing for humanity.

lazygoose 2y ago

Curious about the AI research tool you mentioned, would you mind sharing it? Been trying to get a good local research setup with Ollama but still figuring out what works best.

SamPatt 2y ago
https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ollama https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ol...
Bilal_io 2y ago
Not OP, but based on their mention of Ollama, I can tell you that it has built in search tools, all you need to do is supply an API to one of the tools, or even run one of the search tools locally using docker.

divan 2y ago

Let's revisit this comment in one year – after the explosion of agentic systems. (:

wokwokwok 2y ago
We already have agentic systems; they're not particularly impressive (1). There's no specific reason to expect them to get better. Things that will shift the status quo are: MCST-LLMs (like with ARC-AGI) and Much Bigger LLMs (like GPT-5, if they ever turn up) or some completely novel architecture. [1] - It's provable; if just chaining LLMs are a particular size into agentic systems could scale indefinitely, then you could use a 1-param LLM and get AGI. You can't. QED. Chaining LLMs with agentic systems has a capped maximum level of function which we basically already see with the current LLMs. ie. Adding 'agentic' to your system has a finite, probably already reached, upper bound of value.
- NitpickLawyer 2y ago
  > It's provable; if just chaining LLMs are a particular size into agentic systems could scale indefinitely, then you could use a 1-param LLM and get AGI. You can't. QED. Perhaps I missunderstand your reply, but that has not been my experience at all. There are 3 types of "agentic" behaviour that has worked for a while for me, and I don't know how else it would work without "agents": 1. Task decomposition - this was my manual flow since pre-chatgpt models: a) provide an overview of topic x with chapter names; b) expand on chapter 1 ... n ; c) make a summary of each chapter; d) make an introduction based on the summaries. I now have an "agent" that does that w/ minimal scripting and no "libraries". Just pure python control loop. This gets me pretty reasonable documents for my daily needs. 2. tool use (search, db queries, API hits). I don't know how you'd use an LLM without this functionality. And chaining them into flows absolutely works. 3. coding. I use the following "flow" -> input a paragraph or 2 about what I want, send that + some embedding-based context from the codebase to an LLM (3.5 or 4o, recently o1 or gemini) -> get code -> run code -> /terminal if error -> paste results -> re-iterate if needed. This flow really works today, especially with 3.5. In my testing it needs somewhere under 3 "iterations" to "get" what's needed in more than 80% of the cases. I intervene in the rest of 20%.

atonse 2y ago

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 points to then start googling specific terms, people, laws, treaties, etc to dig deeper and verify. Last year during a visit to my first Indian reservation, I had a whole bunch of questions that nobody in person had answers to. And ChatGPT was invaluable in understanding concepts like where a reservation’s autonomy begins and ends. And why certain tribes are richer than others. What happens when someone calls 911 on a reservation. Or speeds. Or wants to start a factory without worrying about import/export rules. And what causes some tribes to lose their language faster than others. And 20 other questions like this. And most of those resulted in google searches to verify the information. But I literally could never do this before. Same this year when I’m visiting family in India. To learn about the politics, the major players, WHY they are considered major players (like the Chief Minister of Bengal or Uttar Pradesh or Maharashtra being major players because of their populations and economies). Criticisms, explanations of laws, etc etc. For insanely curious people who often feel unsatisfied with the answers given by those around them, it’s the greatest thing ever.

stravant 2y ago
One of my favorite successes was getting an LLM to write me a program to graph how I subjectively feel the heat of steam coming off of the noodles I'm pouring the water out from as a function of the ambient temperature. I was wondering which effects were at play and the graph matched my subjective experience well.
1209412comb 2y ago
Semi-related but I find that sometime it just completely ruined a type of conversation. Like as in your example, I would previously asked people "how would 911 handle an US Reservation Area", and watch how my friends think and reason. To me getting a conclusive answer was not a point. Now they just copy & paste Chat GPT, no fun haha.

bloppe 2y ago

At this point I think even the most bearish have to concede that LLM's are an amazing tool. But OpenAI was never supposed to be about creating tools. They're supposed to create something that can completely take over entire projects for you, not just something that can help you work on the projects faster. If they can't pull that off in the next year or two, they're gonna seriously struggle to raise the next 10B they'll need to keep the lights on. Of course LLMs aren't going anywhere, but I do not envy Sam Altman right now.

lumost 2y ago
At this point it’s quite likely that they could pivot and just be the chatgpt company. I’ve found chatgpt-4o with web search and plugins to be more useful than o1 for most tasks. It’s possible we’re nearing the end of the LLM race, but I doubt that’s the end of the AI story this decade, or OpenAI.
- bloppe 2y ago
  Ya I think they probably will, but "the chatgpt company" is not worth 157B. It might not even be worth 1B.
  - josu 2y ago
    It has replaced ~50% of my Google searches.
    - silisili 2y ago
      Yes but it also hasn't been attacked by ads yet. Google doesn't suck for lack of search results, it sucks because of ads. Imagine asking chatgpt to tell you about slopes in Colorado, and the first five answers are about how awesome North Face is and how you can order from them. You probably wouldn't use it as much.
      - jeffhuys 2y ago
        Local models are GOOD as well, and easy to use (ollama + open web ui). OpenAI has to perform a huge trick in order to stay relevant.

throwup238 2y ago

Amen. Everyone is talking about plateaus and diminishing returns on training but I don’t care one bit. I get that this is a startup focused forum and the financial sustainability of the market players is important but I can’t wait to see what the next decade of UX improvements will be like even if model improvements slow to a crawl.

uludag 2y ago

I'm pretty sure the plan has never been to just make these tools that make us more efficient. If AI stays at the level it's at, it would be a profound failure for companies like OpenAI. We're all benefiting from the capital being poured into these technologies now. The enshittification will come. The enshittification always comes.

mitemte 2y ago
I’m paying $240 a year to Anthropic that I wasn’t paying before and it’s worth it. While I don’t use Claude every single day, but I use it several times a day when I’m working. More times than the free tier allows.
- aprilthird2021 2y ago
  Why do people say this like it's a refutation? Current valuation and investments were not based on getting a very small group of nerds (affectionately) on HN to pay $250/yr which probably doesn't cover even inference costs for the models let alone training and R&D

spaceman_2020 2y ago

This is me. If things never improve and Sonnet 3.6 is the best we have…I’m fine. Its good enough to drastically improve productivity

jayseattle 2y ago

[dead]

BOOSTERHIDROGEN 2y ago

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

alickz 2y ago

i feel like AGI is an arbitrary line in the sand anyway i think as humans we put too much emphasis on what intelligence means relative to ourselves, instead of relative to nature

rkagerer 2y ago

completely local AI research tool, based on Ollama Could you elaborate? Was it easy to install?

arcanemachiner 2y ago
Not OP, but yeah, ollama is super easy to install. I just installed the Docker version and created a little wrapper script which starts and stops the container. Installing different models is trivial. I think I already had CUDA set up, not sure if that made a difference. But it's quick and easy. Set it up, fuck around for an hour or so while you get things working, then you've got your own local LLM you can spin up whenever you want.
- foobiekr 2y ago
  Does ollama still execute whatever arbitrary python code is in the model?
SamPatt 2y ago
Yes I was referring to this: https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ollama https://github.com/TheBlewish/Automated-AI-Web-Researcher-Ol...

htrp 2y ago

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

tugu77 2y ago

Thats all fine, but I think you are missing the bigger picture. It's not about whether what we already got out of this is good. Of course it is. But this is about where it's going. Until about 120 years ago, people were happy with horses and horse carriages. Such a great help! Travel long distances, pull weights, I never want to go back! But then the automobile was invented and within a few years little travel was done by horses anymore. More recently, everybody had a landline phone at home. Such great tech! Talk to grandma hundreds of miles away! I never want to go back! Then suddenly the mobile phone and just shortly after the smart phone came along and now nobody has a landline anymore but everybody can record tiktoks anywhere anytime and share them with the world within seconds. Now imagine "AI". Sure, we have some new tools right now. Sure we don't want to go back. But imagine the transformative effects that could come if the train didn't stop here. Question is just: will it?

hamilyon2 2y ago

If the progress in capabilities stall, the product fit, adoption, ease of use are the next battlefield. OpenAI may be first to realize and switch, so they still have a chance to recoup some of those billions

badgersnake 2y ago

They’re garbage, they will always be garbage. Changing a 4 to a 5 will not make it not garbage. The whole sector is a hype bubble artificially inflating stock prices.

Elextric 2y ago
https://www.youtube.com/watch?v=lFc1jxLHhyM https://www.youtube.com/watch?v=lFc1jxLHhyM
- badgersnake 2y ago
  If that’s supposed to be impressive, it really isn’t.

ninetyninenine 2y ago

The definition of agi is a linguistic problem but people confuse it for a philosophical problem. Think about it. The term is basically just a classification and what features and qualities fit the classification is an arbitrary and linguistic choice. The debate stems from a delusion and failure to realize that people are simply picking and choosing different fringe features on what qualifies as agi. Additionally the term exists in a fuzzy state inside our minds as well. It’s not that the concept is profound. It’s that some of the features that define the classification of the term we aren’t sure about. But this doesn’t matter because we are basically just unsure about the definition of a term that we completely made up arbitrarily. For example the definition of consciousness seems like a profound debate but it’s not. The word consciousness is a human invention and the definition is vague because we choose the definition to be ill defined, vague and controversial. Much of the debate on this stuff is purely as I stated just a language issue.

zifpanachr23 2y ago
If it's genuinely what you say, then how is what is going on not slavery? I don't believe AGI is possible but if it was and it was as subjective as you say what is and isn't conscious, then it starts to take on an even more altogether evil character. Akin to cloning slave humans or something for free cheap labor.
- ninetyninenine 2y ago
  how does a linguistic and language issue relate to slavery. It's the definition of a word. That's all. Slavery is also a word. Don’t you find it strange that your entire moral framework is constructed on top of the basis of arbitrary definitions of vocabulary? Make what you think is right or wrong based not off of language. Language is a delusion that masquerades as something with actual meaning when it is just an invention, a tool, to facilitate communication. Right now your concept of right and wrong is a vocabulary issue. Does this make sense? No.

delusional 2y ago

As a consumer you should always evaluate the product that is in front of you, not the one they promise in 6 months. If what's there is valuable to you, then that's great. When we discuss the potential AGI we're not talking as consumers, we're talking about the business side. If AGI is not reached, you'll see an absolutely enormous market correction, as it realizes that the product is not going to replace any human workers. The current generation of products are not profitable. They're investments towards that AGI dream. If that dream doesn't happen, then the current generation of stuff will disappear too, as it becomes impossible to provide at a cost you'd be comfortable with.

rajamaka 2y ago
Human workers workers have already been replaced.

ryukoposting 2y ago

At this point, most conceivable beneficial use cases for LLMs have been covered. If the economics of AI tech were aligned with making a good product that people want and/or need, we'd basically take everything we have at this point and make it lighter, smaller, and faster. I doubt that's what will happen.

michaelbuckbee 2y ago

What was the AI search tool?

tiffanyh 2y ago

I have the opposite reaction. AI right now feels like that MBA person at work. They don’t know anything. But because they sound like they are speaking with authority & confidence, allows them to get promoted at work. (While all of the experts at work roll their eyes because they know the MBA/AI is just spitting out nonsense & wish the company never had any MBA/AI people)

zppln 2y ago
And the MBA person (at my company this is everyone in middle management) is also the person who go around and suggest we shoehorn AI into everything...

fud101 2y ago

how do you interact with perplexity? mobile app?

ByteAndBattle 2y ago

Local search with Ollama? Please share!