Pressing the copilot button to instantly bring up a text box where you can interact with an LLM is amazing UI/UX for productivity. LLMs are by far the best way to retrieve information(that doesnt need to be correct).
If this had been released with Agentic features that allow it to search the web, use toolscripts like fetching time/date and stuff from the OS, use recall, properly integrate with the microsoft app suite. It would be game changing.
We already have proof that this is a popular feature for users since its been integrated in every mobile phone for the past 10 years.
I am so confused as to how the majority of people view this community.
This post right now shows -12 vote count. So does that mean this is a popular opinion…? Or do the majority of people not know how this works?
Because users can smell a shill post a mile away
It could be traffic from outside this community, so they havent read the rules in the sidebar. Maybe people just hate copilot so much its a subconscious reaction lol.
LLM is just a slow way to do things that have better ways to do them.
Or to have an expensive autocorrect do your thinking.
Upvoted. It’s utterly useless.
So you agree that pressing a button to bring up a box that you can query with natural language is a good feature you just think the LLM part is slower and computationally inefficient? I could agree with that if there was something better proposed. I just see an LLM being a good tech for this because of how dynamic it is and with the addition of tools to do specific tasks in a determinism fashion its a powerful tool for the users.
Clearly you haven’t worked with one.
Its great for getting detailed references on code, or finding sources for info that would take a LOT longer otherwise.
Not who you’re responding to, but I used one extensively in a recent work project. It was a matter of necessity, as I didn’t know how to word my question in the technical terms specific to the product, and it was something that was just perfect for search engines to go “I think you actually mean this completely different thing”. There was also a looming deadline.
Being able to search using natural language, especially when you know conceptually what you’re lookong for but not the product or system specific technical term, is useful.
Being able to get disparate information that is related to your issue but spread across multiple pages of documentation in one spot is good too.
But detailed references on code? Reliable sources?
I have extensive technical background. I had a middling amount of background in the systems of this project, but no experience with the specific aspects this project touched. I had to double check every answer it gave me due to how critical what I was working on was.
Every single response I got had a significant error, oversight, or massive concealed footgun. Some were resolved by further prompting. Most were resolved by me using my own knowledge to work from what it gave me back to things I could search on my own, and then find ways to non-destructively confirm the information or poke around in it myself.
Maybe I didn’t prompt it right. Maybe the LLM I used wasn’t the best choice for my needs.
But I find the attitude of singing praises without massive fucking warnings and caveats to be highly dangerous.
Great response.
It’s great until you realize it’s led you down the garden path and the stuff it’s telling you about doesn’t exist.
It’s horrendously untrustworthy.
I’ve spent many many hours working with LLMs to produce code. Actually, it’s an addictive loop, like pulling a slot machine. You forget what you’re actually trying to accomplish, you just need the code to work. It’s kinda scary. But the deeper you get, the worse the code gets. And eventually you realize, the LLM doesn’t know what it’s talking about. Not sometimes, ever.
It has been useful for me with poorly documented libraries, not generating more than code snippets or maybe small utilities.
It’s more of an API search engine to me. I find it’s about 80% correct but it’s easier to search for a specific method to make sure it does what you expect than scroll through pages of generated class documentation, half of which look like internal implementation details I won’t need to care about unless I’m really digging into it as a power user.
Also, even if the method isn’t correct or is more convoluted to use than a more direct one. it’s usually in the same module as the correct one.
That’s some funny shit.
Clearly you’ve not been fact checking the shit it hallucinates.
Skill issue. I’m better at retrieving and then actioning real and pertinent information than you and an AI combined, guaranteed.
or finding sources for info that would take a LOT longer otherwise.
Maybe. It adds to the list of sources you have to check from, but i’ve found i still have to manually check to see if it’s actually on topic rqther than only tangentially related to what I’m writing about. But that’s fair enough, because otherwise it’d be like cheating, having whole essays written for you.
Its great for getting detailed references on code
I know it’s perhaps unreasonable to ask, but if you can share examples/anecdotes of this I’d like to see. To understand better how people are utilising LLMs
Upvoted for absolutely horrendous take.
Keshee out in the wild? Upvote!
LLMs are by far the best way to retrieve information(that doesnt need to be correct).
What is the point in retrieving information if it isn’t correct?
To lower the general populations IQ and make it easier to hide misinformation and propaganda inside of Ai
specifically NEED. Very few things in our day to day life NEED to be correct. Typically its good to be correct but our usecases can handle being wrong because its either low stakes or we will be diving deeper into the topic as we narrow down our information search.
We do these kinds of searches all the time. Everytime you ask an average person a question you’re preforming one of these searches. Everytime something pops into your head and you want a quick answer you’re preforming one of these searches. When you search for information online you’re generally preforming one of these kinds of searches.
An example could be I want to know a few of the popular python libs for interacting with atlassian. It gives me a list of some libs and links and I can go check them out.
I would much rather be directed to correct information than be told information that may not be correct. Bad information causes me to waste my time and money.
If its a situation where bad info can cause you to waste significant time and lose money then Copilot is not the right tool for the job. But thats not every query and I wouldnt even say its more than half our search queries.
Its boring that people are completely unwilling to engage at all with the arugment. Comparing a situation where you get the wrong info from an LLM with a situation where you get the correct info from google is pointless and shows nothing. Ignoring that people get the wrong info from google all the time is so disingenuous. Everyone would acknowledge google is still a useful resource even if there is the chance to come away with the wrong answer.
So the real comparison to make isnt correct answer from google 100% of the time vs correct answer from copilot 60% of the time. Its correct answer from google 90% of the time vs copilots 70%. So you can weigh up the options and use the right tool for the job.
retrieve information(that doesnt need to be correct).
Perhaps I’m just one of “the olds” who doesn’t get modern technology, and this is why I’m having a real difficult time imagining why I, or anyone, would ever spend time looking something up when the factual correctness is optional to begin with.
Yeah, if I don’t care about correctness, i can just make it up myself
Unless you think people always come away from google with the right answer I dont see the 1:1.
If you NEED the right answer you should go to a trusted source same as if you’re using google. If you are looking for an answer then usually blogspam articles, reddit, or AI will all be good enough to return something satisfying. AI is just a faster way of searching a question on google and clicking thte top result.
a faster way of searching a question on google and clicking thte top result.
No, it isn’t. The “I’m feeling lucky” button is.
No its not. Firstly 99% of people have no idea what that button is.
Secondly opening a web browser and going to google typing in your question then pressing ‘im feeling lucky’ then searching through the webpage is way slower than hitting the copilot button typing your question and getting a quick direct answer.
Then write yourself a desktop plugin, an icon, an input box, anything, to take you to the first Google search result. What the fuck does this have to do with LLM? How is this justified to use gallons of water, gigawatt of electricity, and PBs of stolen training data?
Ok so your main complaint is that its to energy intensive? Would you concede that its an OS assistant is a good feature if the query computation cost was lowered? Because I’d argue it already is and the cost of an LLM query isnt unreasonable. The large power costs come from model training and per query cost is negligible.
Also I wont make an argument on the copyright for training data because i dont respect copyright.
upvoting because this is a good unpopular opinion.
unfortunately, microsoft is about ten years too early to the party, like they always are. what they offer isnt very reliable either, in my opinion
-
No it’s not
-
No it won’t. Not even becsuse it’s shit, but becsuse it’s Microsoft’s and they sure as fuck aren’t gonna put it in any OS other than Windows.
-
It has not been added to every phone and isn’t even 10 fucking years old. It’s barely existed for 2.
Are you just using “Copilot,” a specific brand of generative AI assistant program like the way a parent would call all video games “Nintendo?”
3. It has not been added to every phone and isn’t
even 10 fucking years old. It’s barely existed for 2.
It might be that they are conflating GenAI with services such as Google Assistant ot Siri. Though I personally find that Google Assistant is/was more useful that their GenAI implementation (Gemini).
Both siri and siri with an LLM is the same product its just an expanded scope due to the addition of new tech. I see no reason why i cant group them.
-
Obviously I am not talking about copilot being built in to other OSs. I am talking about an OS assistant that uses LLMs to handle natural language queries being built into OSs.
-
Yes it has, IOS and Android have had siri, bixby and google assistant for over a decade.
In copilots current form its chatgpt with some basic functionality added on. Its a baby compared to siri, bixby and google assistant. I think its a good addtition to windows and we will see it grow to be a more powerful version of google assistant and we will see this develop and be implemented on the other major desktop OSs.
-
-
The fuck it is. People will make uninstallers or blockers.
Thats fine, I dont expect 100% of people to like anything. My argument is a general one.
Sometimes I wonder if people come here and derive ludicrous drivel for the explicit purpose of posting here.
This is one of those times.
Its a slow day at work, I wont lie.
I’m not here to upvote, downvote, or argue, but I do not use this feature, desktop, mobile or otherwise, I don’t even use the web pages like an ancient Internet user, LLMs only provide productivity when you already know what you’re doing and you’re basically guiding a toddler through a laser field, using the chat features for anything besides coming up with clever names you might name your next dog is basically never going to increase your productivity.
I disagree with your overall opinion for various reasons (relying on AI erodes researching and critical thinking skills; Copilot is dangerously unreliable in the majority of use cases); it is invasive to the point that it’s creating a user backlash; there’s are many serious social/emotional issues that are surfacing because of AI over-use; etc etc).
But I respect that you have shared a genuinely unpopular opinion here (in the right community). And you put your arguments forward in a well-worded and coherent way. So kudos.
ETA: I don’t think it’s appropriate to personally insult OP because of this post, as a small number of people here are doing. C’mon people, look at the community we are in and don’t resort to such insults. This is a great topic for legit discussion.
I agree that LLMs can erodes critical thinking skills and can be unreliable but I think they ARE fit for purpose for majority of searches and queries people have day to day and people are figuring out what is and isnt a question for an LLM. Like im not going to ask an LLM how to configure some piece of software I’ll go to the docs and read it because i need this to be configured correctly. I wouldnt ask an LLM if I can eat this weird mushroom because i might die if its wrong.
But I would ask an LLM what tech I can use if I want to get X result and then look through the summaries of each suggestion. I would ask for a report or document template to be generated because im proof reading the document anyway. I would ask for help automating a task. I would ask for help writing random low effort slop posts that I have to do for office stuff, like marketing emails, event announcements etc.
My reasoning for this post is that even though I dont like copilot currently I can see that at its core its a good feature and with the right polish it can be a great improvement for users. A big gripe i have is that marketers have way overpromised what assistants like copilot can do. When i speak with other people I can see they have already been leaning heavily on natural language queries for over a decade now and having this built into the OS would be a huge quality of life improvement and would improve what the tool can do. People already been outsourcing their thinking to google many years ago so I cant pearl clutch over doing the same with chatgpt and we can put our heads in the sand(like most people in this thread) and pretend people arent using these LLMs for information but the reality is that they are and we need to accept it and be involved in building the software that people want.
I’m a linux user and I think it would be very useful to be able to click one button and say “Set a calander event for the 25th my dads birthday and set a reminder a week earlier” and have it set that. It works on mobile just fine.
I don’t know why people are downvoting this. It really is an unpopular opinion.
just sell your microslop shares, this isn’t happening
since its been integrated in every mobile phone for the past 10 years.
We just found out the person who makes job announcements asking for more years of experience with a technology than the technology’s years in existence
Do you not think assistances have been in phones for 10 years or are you saying they are different specifically because they didnt use LLMs back then?
It was just a silly joke about llms being much more recent :D. The way you phrased it made it sound like you were talking just about llms, and not software assistants in a broader sense
every phone
Not every one. I have GrapheneOS on the phone. And Linux on the computer. Like some nerds here. And there hasn’t been any assistant popping up on mine…
Also don’t think whether that’s going to be built into any major operating system is much of an opinion. That’s more a fact 😉But yeah, whether that’s useful… I’ve read all kinds of opinions on that. I think we need more factual data on user efficiency. I’m positive we’ll get some more studies on that.
Not every one. I have GrapheneOS on the phone.
ah yes i forgot to list the 100 edge cases that cover 0.0001% of the phone market but you know what I mean.
I think we will get something similar to copilot on linux, obviously there will be distros that dont ship it but people are working on projects and once the open source models improve and the UI/UX workflows get polished up I believe major distro will ship them. The linux foundation owns the MCP protocol which is the goto protocol for AI’s to interact with things.
I also doubt they provide a huge increase in productivity. I’d say between 0% and 10% productivty increase and its more of a quality of life and small time saver than a powerhouse productivty tool like its marketing suggests.
Phones, sure. I -personally- think it’s massively problematic how the phone operating system ecosystem is basically a monopoly of two companies. And I almost can’t do my paperwork or get a doctor’s appointment or train ticket anymore without accepting to forward my personal information to a list of 40 “partner” companies, a good chunk of them abroad in the USA. And then it’s massively complicated and I need 3 authenticator apps, and they do device verification and SafetyNet to make sure my(?) phone isn’t controlled by me, but Google.
So yes, in reality it’s not how I envision it to be. Phones just do what Google wants them to do and that certainly also includes Gemini AI. All of this is almost impossible to avoid, and it’s getting harder each day. It certainly is that way.
(Same with edge-cases in general. I had to contact modern customer support lately, and that just got way worse than it already was before AI chatbots. We just don’t do edge-cases any more. Everyone needs to get in line, have the same life and same common issues or they’re screwed.)
With Linux, I doubt it. Traditionally it’s a lot about choice. Caters to its user group who (on the desktop) include a good amount of privacy advocates, people with older computers, nerds… I think we’d need some paradigm shift first. Before any of the larger distributions change their defaults. I like to believe we’re relatively safe here. And my biggest issue isn’t AI in itself, but how large, annoying companies shove it down our throats. And that’s really not how Linux works.
I’ve been pondering productivity as well. I once did some AI assisted coding. Took notes and did a similar task after that the old-school way. In that case AI had wasted time, I was faster without. But it’s been a while and AI tools have improved in the last months. So I probably should repeat that experiment. And do it a couple times to get some solid numbers. I find it hard to apply it the exact right way, though. It’d underperform (on me) if I don’t get the prompt right, feed it the right amount of context if there’s a pre-existing project… It’s better at some tasks and not so good at other ones… So with the current state of technology it’s not that straightforward to delegate stuff to AI, and it’d just increase productivity. At least that’s been my previous experience. But we get a plethora of contradicting and weird reports on AI’s performance when used for coding.
