Rendered at 00:50:20 GMT+0000 (Coordinated Universal Time) with Cloudflare Workers.
Mabusto 1 days ago [-]
As long as I can connect my 3D TV, I can't wait to sip some Juicero and watch Quibi on it.
Avicebron 1 days ago [-]
As long as the models are local this doesn't seem that crazy. What concerns me is that these are "agentic PCs" that only work with a subscription.
Animats 1 days ago [-]
The "agentic PC" for consumers probably would be something you talk to, and would look like Alexa or a living room TV or glasshole glasses. Something other than a keyboard and screen combo.
devn0ll 1 days ago [-]
Seeing how we are responding to AI or the copilot button on pc's... I _dare_ to suggest this "Agentic era" is nothing more then wishful thinking. Not supported by any real wish or need on the consumer side.
Well, at least for me: no thanks.
cromka 1 days ago [-]
As much as I hate it, you and your opinion is exactly what Jobs always talked about: people don't need things until they realize they need them.
You not envisioning use for it is just a past bias. You can't know that. You can't because we haven't yet reached the point where the OS is fully useful when controlled with AI.
gedy 24 hours ago [-]
But the industry types have been talking about "agents" for 30 years... This used to be a thing that "intelligent agents will go on the internet and gather information for you" then search engines came out and people were happy with that instead.
spankibalt 21 hours ago [-]
Yeah, just a week ago I read an interview with Alan Kay from 1990, where he shared his thoughts about the third revolution in computing: the "intimate computer" (the agentic platform that follows the "institutional computer" and the "personal computer").
cromka 17 hours ago [-]
Except it is only now when this is actually possible.
spankibalt 13 hours ago [-]
Funnily enough, he spoke about one of his first "agentic computing" implementations: a computer agent of his he tasked with compiling a sort of newspaper for breakfast reading. It ran overnight for 12 to 15 hrs., and collated textual as well as graphical information from specified news resources and databases. That was ten years before the interview, in 1980. Sadly, the piece doesn't go into more detail on the setup or its performance metrics.
He also mentioned that the idea of agentic computing was already 30 years old, and that he was busying himself with the topic for 15 years by then (1990). So... five years from taking interest (mid-70s) to his first practical implementations (1980).
adrian_b 14 hours ago [-]
The idea about various fads that "not envisioning use for it is just a past bias", is seldom true.
What people want has not changed for millennia and it is unlikely to change soon.
Most of the things that are useful have already been imagined millennia ago, even if at that time nobody had any idea about how one could develop any technology for building such things in reality. For instance in the Ancient Greek literature there are descriptions of artificial robots for doing the hard work, means for flying etc.
The past bias can block indeed one to envision the usefulness of some things, but only when those things are not a goal in themselves, but they are only intermediates for achieving things that are already known to be useful and the past bias prevents the user to realize that there exists an alternate path to the useful goals, instead of the known traditional path.
LLMs are indeed tools that can be used to achieved some useful goals, so in some cases a user may not realize how they can be used, due to past bias.
There is no doubt that there are a few applications for which LLMs are very useful, but for experienced people, even if they have never used LLMs yet, it is easy to recognize with certainty that some of the proposed applications for LLMs will never be useful for them.
For example, I would never use an LLM for searching the Web or for summarizing documents. What I recognize as important in a Web search or in a document differs too much from what typical humans would recognize, for an LLM to have any chance to generate equivalent results.
The only reason why I may find useful to put some questions to a big LLM is because it is likely that it may have had access during training to documents to which I do not have access. Thus the answer might provide some clues about other sources than those known to me. Instead of this, I would very much prefer to use a traditional search tool on the training set, but the LLM may be a poor substitute for its training set, which is better than nothing.
For now, the most lucrative application for LLMs is as coding assistants. Here there is no past bias, because since the earliest times of automatic computers, people have hoped for methods that would allow the generation of computer programs with minimum input from a human.
I do not think that there is anyone who would dispute that LLMs have allowed a much greater progress than before in this direction. Here what are frequently disputed are only the correct strategies of using LLMs for this purpose, because it is obvious that they are frequently misused and those who do not understand programming, like most managers, have completely unrealistic ideas about what can be done and what should be done with LLMs.
sublinear 23 hours ago [-]
I'm sorry, but what was Steve Jobs ever uniquely right about except that we needed better touch screens on smartphones?
We don't see the same obvious applications of AI because nobody has developed a proper user interface for it. We're stuck with voice, chat, and dumping documents onto it. The current pro-AI stance is basically "fuck the user and fuck interfaces".
Closi 1 days ago [-]
> Seeing how we are responding to AI or the copilot button on pc's... I _dare_ to suggest this "Agentic era" is nothing more then wishful thinking
Depends who 'we' is - I've seen plenty of non-tech people in the real world begin to use ChatGPT as a primary information source rather than the web (rightfully or not!)
I suspect that 'we' might not be the true early adopters here, similar to how quite a lot of the most technical users in the 80's thought GUI's were a waste of time.
Octoth0rpe 1 days ago [-]
> Depends who 'we' is - I've seen plenty of non-tech people in the real world begin to use ChatGPT as a primary information source rather than the web (rightfully or not!)
I don't think that's really what people are talking about when they talk about 'agentic' PCs.
Closi 18 hours ago [-]
I think we are more broadly talking about what the user interface is going to be in the future and how people will interact with a computer - many people already want to interact with ChatGPT to get answers rather than navigate to a website, or want to prompt ChatGPT to generate a leaflet rather than design one in Microsoft Word, so 'Agentic PCs' is just an extension to that.
cromka 1 days ago [-]
I think that was example of how previously seemingly impossible things are happening things are happening quickly.
tmaly 1 days ago [-]
I had to buy a new washing machine last year. It has an AI mode, what ever that is. I have never used the mode.
syberspace 1 days ago [-]
as far as I've understood the AI mode on my new-ish washing machine: it's just a renamed "automatic" mode that uses a sensor to measure how heavy the load is and adjusts the cycle length. there is absolutely no AI involved, just an if-statement or equivalent logic gates. I'd guess yours does something similar
Animats 1 days ago [-]
Washers now do have useful control systems. Mine starts out by spinning the tub a little, before adding water, to measure the load. Out of balance problems are a thing of the past - that's sensed and dealt with automatically. It's able to handle bed comforters or sneakers without problems. But it's not "AI", and it doesn't have a network connection.
jagged-chisel 1 days ago [-]
Literally, “simulated intelligence” at best.
But for marketing, “artificial intelligence” is fine. And better than LLMs being called “AI”
ajam1507 1 days ago [-]
Surprising that there are still people who don't think LLMs qualify as AI
Octoth0rpe 1 days ago [-]
I think in some people's minds, the concept of sentience and intelligence are intertwined, and there are at least some people (myself included) who do not think they're the same. There is a strong (but surprisingly not universal) consensus that LLMs are not sentient, so if you insist that sentience/intelligence are the same thing, then LLMs don't qualify as AI either. If you think the two concepts are separable, then they're intelligent but not sentient. The devil is of course in the definitions.
squid_ca 24 hours ago [-]
Mine too. I think it's just a buzzword, like, this would have been called "Smart Wash" five years ago.
andai 1 days ago [-]
On my walk today I passed an LG company van. It had an ad for one of their new AC units on the side. "AI Air"
throw-the-towel 22 hours ago [-]
A bar I know had an "AI designed shot" back in 2023.
forinti 1 days ago [-]
There's a billboard near my home with an ad for "AI designed glasses".
1 days ago [-]
georgeburdell 22 hours ago [-]
Agentic for most people should mean an easy way to fix what were previously missing UI features. Hopefully companies are (with permission) paying attention to user prompts in aggregate to guide UX work
corv 1 days ago [-]
Less space than a nomad. Lame.
To be fair, I find the term to be as contrived as “performant”
XorNot 1 days ago [-]
What you're not planning to upgrade from your Web 3.0 platform to an Agentic one?
Scandalous!
Henchman21 1 days ago [-]
Given that we’re making ourselves dumber through AI, education funding cuts, social media, and foolish propaganda AND the population is shrinking and everyone is seemingly depressed:
The most likely outcome is the world in the children’s cartoon “Thundarr the Barbarian”. People living in the collapsed ruins of the past society, belief in magic, etc.
A post-apocalyptic hellscape, essentially.
andai 1 days ago [-]
> People living in the collapsed ruins of the past society
Kind of feels like I was already born into that.
baron3dl 24 hours ago [-]
is this not an essential human condition?
yogthos 1 days ago [-]
I think it's almost certain that we'll be moving to running local models as a default in a few years. The quality of small models has been improving at an astonishing rate in my opinion. My favorite example is how Qwen3.6-27B that you can run on a laptop outperforms Qwen3.5-397B which was a flagship model requiring a commercial grade server that was released just in February. https://qwen.ai/blog?id=qwen3.6-27b
I fully expect that local models models that are comparable to current frontier models in performance will appear in the near future. Additionally, a lot more can be done with the harness as well, which in my opinion is an under-explored territory right now. For example, ATLAS does some clever tricks in this area https://github.com/itigges22/ATLAS
I started working on my own harness and also notice a significant improvement in model capability with it https://dirge-code.github.io
Apple seems to be one of the few companies to have realized that the future is likely local, and they've been focusing on optimizing hardware for that while everybody else seems to still be stuck in a model as a service paradigm.
baron3dl 24 hours ago [-]
I think Apple's tech-heavy user base and vertically integrated hardware/network/software mega architecture positioned them perfectly to beat the rest of the market to 1st runner up. The competition knows, they just can't move that fast.
> I started working on my own harness and also notice a significant improvement in model capability with it https://dirge-code.github.io
You should mine your session logs for examples of scenarios that demonstrate this improvement. If you can characterize it in a time series metric, like tokens/feature, as you applied improvements, then you're offering a receipt.
yogthos 23 hours ago [-]
Yeah, that's a good idea. I haven't really been rigorous with tracking the token usage metrics when I started. I was thinking I could compare solving tasks with opencode too and track metrics for both.
Well, at least for me: no thanks.
You not envisioning use for it is just a past bias. You can't know that. You can't because we haven't yet reached the point where the OS is fully useful when controlled with AI.
He also mentioned that the idea of agentic computing was already 30 years old, and that he was busying himself with the topic for 15 years by then (1990). So... five years from taking interest (mid-70s) to his first practical implementations (1980).
What people want has not changed for millennia and it is unlikely to change soon.
Most of the things that are useful have already been imagined millennia ago, even if at that time nobody had any idea about how one could develop any technology for building such things in reality. For instance in the Ancient Greek literature there are descriptions of artificial robots for doing the hard work, means for flying etc.
The past bias can block indeed one to envision the usefulness of some things, but only when those things are not a goal in themselves, but they are only intermediates for achieving things that are already known to be useful and the past bias prevents the user to realize that there exists an alternate path to the useful goals, instead of the known traditional path.
LLMs are indeed tools that can be used to achieved some useful goals, so in some cases a user may not realize how they can be used, due to past bias.
There is no doubt that there are a few applications for which LLMs are very useful, but for experienced people, even if they have never used LLMs yet, it is easy to recognize with certainty that some of the proposed applications for LLMs will never be useful for them.
For example, I would never use an LLM for searching the Web or for summarizing documents. What I recognize as important in a Web search or in a document differs too much from what typical humans would recognize, for an LLM to have any chance to generate equivalent results.
The only reason why I may find useful to put some questions to a big LLM is because it is likely that it may have had access during training to documents to which I do not have access. Thus the answer might provide some clues about other sources than those known to me. Instead of this, I would very much prefer to use a traditional search tool on the training set, but the LLM may be a poor substitute for its training set, which is better than nothing.
For now, the most lucrative application for LLMs is as coding assistants. Here there is no past bias, because since the earliest times of automatic computers, people have hoped for methods that would allow the generation of computer programs with minimum input from a human.
I do not think that there is anyone who would dispute that LLMs have allowed a much greater progress than before in this direction. Here what are frequently disputed are only the correct strategies of using LLMs for this purpose, because it is obvious that they are frequently misused and those who do not understand programming, like most managers, have completely unrealistic ideas about what can be done and what should be done with LLMs.
We don't see the same obvious applications of AI because nobody has developed a proper user interface for it. We're stuck with voice, chat, and dumping documents onto it. The current pro-AI stance is basically "fuck the user and fuck interfaces".
Depends who 'we' is - I've seen plenty of non-tech people in the real world begin to use ChatGPT as a primary information source rather than the web (rightfully or not!)
I suspect that 'we' might not be the true early adopters here, similar to how quite a lot of the most technical users in the 80's thought GUI's were a waste of time.
I don't think that's really what people are talking about when they talk about 'agentic' PCs.
But for marketing, “artificial intelligence” is fine. And better than LLMs being called “AI”
To be fair, I find the term to be as contrived as “performant”
Scandalous!
The most likely outcome is the world in the children’s cartoon “Thundarr the Barbarian”. People living in the collapsed ruins of the past society, belief in magic, etc.
A post-apocalyptic hellscape, essentially.
Kind of feels like I was already born into that.
I fully expect that local models models that are comparable to current frontier models in performance will appear in the near future. Additionally, a lot more can be done with the harness as well, which in my opinion is an under-explored territory right now. For example, ATLAS does some clever tricks in this area https://github.com/itigges22/ATLAS
I started working on my own harness and also notice a significant improvement in model capability with it https://dirge-code.github.io
Apple seems to be one of the few companies to have realized that the future is likely local, and they've been focusing on optimizing hardware for that while everybody else seems to still be stuck in a model as a service paradigm.
> I started working on my own harness and also notice a significant improvement in model capability with it https://dirge-code.github.io
You should mine your session logs for examples of scenarios that demonstrate this improvement. If you can characterize it in a time series metric, like tokens/feature, as you applied improvements, then you're offering a receipt.