This is a most excellent place for technology news and articles.
A 100% accurate AI would be useful. A 99.999% accurate AI is in fact useless, because of the damage that one miss might do.
It's like the French say: Add one drop of wine in a barrel of sewage and you get sewage. Add one drop of sewage in a barrel of wine and you get sewage.
I think it largely depends on what kind of AI we're talking about. iOS has had models that let you extract subjects from images for a while now, and that's pretty nifty. Affinity Photo recently got the same feature. Noise cancellation can also be quite useful.
As for LLMs? Fuck off, honestly. My company apparently pays for MS CoPilot, something I only discovered when the garbage popped up the other day. I wrote a few random sentences for it to fix, and the only thing it managed to consistently do was screw the entire text up. Maybe it doesn't handle Swedish? I don't know.
One of the examples I sent to a friend is as follows, but in Swedish;
Microsoft CoPilot is an incredibly poor product. It has a tendency to make up entirely new, nonsensical words, as well as completely mangle the grammar. I really don't understand why we pay for this. It's very disappointing.
And CoPilot was like "yeah, let me fix this for you!"
Microsoft CoPilot is a comedy show without a manuscript. It makes up new nonsense words as though were a word-juggler on circus, and the grammar becomes mang like a bulldzer over a lawn. Why do we pay for this? It is buy a ticket to a show where actosorgets their lines. Entredibly disappointing.
Most AIs struggle with languages other than English, unfortunately, I hate how it reinforces the "defaultness" of English
99.999% accurate would be pretty useful. Theres plenty of misinformation without AI. Nothing and nobody will be perfect.
Trouble is they range from 0-95% accurate depending on the topic and given context while being very confident when they’re wrong.
We're not talking about an AI running a nuclear reactor, this article is about AI assistants on a personal phone. 0.001% failure rates for apps on your phone isn't that insane, and generally the only consequence of those failures would be you need to try a slightly different query. Tools like Alexa or Siri mishear user commands probably more than 0.001% of the time, and yet those tools have absolutely caught on for a significant amount of people.
The issue is that the failure rate of AI is high enough that you have to vet the outputs which typically requires about as much work as doing whatever you wanted the AI to do yourself, and using AI for creative things like art or videos is a fun novelty, but isn't something that you're doing regularly and so your phone trying to promote apps that you only want to use once in a blue moon is annoying. If AI were actually so useful you could query it with anything and 99.999% of the time get back exactly what you wanted, AI would absolutely become much more useful.
People love to make these claims.
Nothing is "100% accurate" to begin with. Humans spew constant FUD and outright malicious misinformation. Just do some googling for anything medical, for example.
So either we acknowledge that everything is already "sewage" and this changes nothing or we acknowledge that people already can find value from searching for answers to questions and they just need to apply critical thought toward whether I_Fucked_your_mom_416 on gamefaqs is a valid source or not.
Which gets to my big issue with most of the "AI Assistant" features. They don't source their information. I am all for not needing to remember the magic incantations to restrict my searches to a single site or use boolean operators when I can instead "ask jeeves" as it were. But I still want the citation of where information was pulled from so I can at least skim it.
99.999% would be fantastic.
90% is not good enough to be a primary feature that discourages inspection (like a naive chatbot).
What we have now is like...I dunno, anywhere from <1% to maybe 80% depending on your use case and definition of accuracy, I guess?
I haven't used Samsung's stuff specifically. Some web search engines do cite their sources, and I find that to be a nice little time-saver. With the prevalence of SEO spam, most results have like one meaningful sentence buried in 10 paragraphs of nonsense. When the AI can effectively extract that tiny morsel of information, it's great.
Ideally, I don't ever want to hear an AI's opinion, and I don't ever want information that's baked into the model from training. I want it to process text with an awareness of complex grammar, syntax, and vocabulary. That's what LLMs are actually good at.
For real. If a human performs task X with 80% accuracy, an AI needs to perform the same task with 80.1% accuracy to be a better choice - not 100%. Furthermore, we should consider how much time it would take for a human to perform the task versus an AI. That difference can justify the loss of accuracy. It all depends on the problem you're trying to solve. With that said, it feels like AI in mobile devices hardly solves any problems.
Perplexity is kinda half-decent with showing its sources, and I do rely on it a lot to get me 50% of the way there, at which point I jump into the suggested sources, do some of my own thinking, and do the other 50% myself.
It's been pretty useful to me so far.
I've realised I don't want complete answers to anything really. Give me a roundabout gist or template, and then tell me where to look for more if I'm interested.
AI is useless and I block it anyway I can.
"Stop trying to make ~~fetch~~ AI happen. It's not going to happen."
AI is worse that adding no value, it is an actual detriment.
I feel like I'm in those years of You really want a 3d TV, right? Right? 3D is what you've been waiting for, right? all over again, but with a different technology.
It will be VR's turn again next.
I admit I'm really rooting for affordable, real-world, daily-use AR though.
A damning result for AI pump and dump scammers.
every NVDA earnings call lol. Old man Jenson had a (chip) farm, AI AI OH! guy literally said AI almost 100 times in a call.
Unless it can be a legit personal assistant, I’m not actually interested. Companies hyped AI way too much.
"useless" is a more positive impression than I have.
Ai is a waste of time for me; I don't want it on my phone , I don't want it on my computer and I block it every time I have the chance. But I might be old fashioned in that I don't like algorithms recommending anything to me either. I never cared what the all seeing machine has to say.
I wonder if it has anything to do with the fact that it’s useless.
I don't think it's meant to be useful....for us, that is. Just another tool to control and brainwash people. I already see a segment of the population trust corporate AI as an authority figure they trust.
How could it not be this way? Algorithms trained people to never seek out information on their own. They're trained to be fed info from the rich and never seek anything out on their own.
Like, we say dead internet. Except...nothing is actually stopping us from ditching corporate internet websites and just go back to smaller privately owned or donation run forums.
Check out debate boards. Full of morons using ChatGPT to speak for them and they'll both openly admit it and get mad at you for calling it dehumanizing and disrespectful.
/tinfoil hat
Edit to add more old man yells at clouds detail.
AI was never meant for the average person but the average person had to be convinced it was for funding.
Nothing bores me more than their events that focus on AI.
The AI thing I'd really like is an on-device classifier that decides with reasonably high reliability whether I would want my phone to interrupt me with a given notification or not. I already don't allow useless notifications, but a message from a friend might be a question about something urgent, or a cat picture.
What I don't want is:
It seems like those are the main AI features bundled on phones now, and I have no use for any of them.
That's useful AI that doesn't take billions of dollars to train, though. (it's also a great idea and I'd be down for it)
The first thing I do with a new phone is turn off any kind of assistance.
Not sure if Google Lens counts as AI, but Circle to Search is a cool feature. And on Samsung specifically there is Smart Select that I occasionally use for text extraction, but I suppose it is just OCR.
From Galaxy AI branded features I have tested only Drawing assist which is an image generator. Fooled around for 5 minutes and have not touched it again. I am using Samsung keyboard and I know it has some kind of text generator thing, but have not even bothered myself to try it.
Certainly counts, Samsung has a few features like grabbing text from images that I found useful.
My problem with them is its all online stuff and I'd like that sort of thing to be processed on device but thats just me.
I think folks often are thinking AI is only the crappy image generation or chat bots they get shoved to. AI is used in a lot of different things, only difference is that those implementations like drawing assist or that text grabbing feature are actually useful and are well done.
Sometimes I wonder what is going to happen to all this tech in 4 or so years when its less profitable to keep the AI centers on.
Right now they are "free" because of all the investment that is going on. But they have a huge maintenance/energy cost.
I'm shocked, I tell you. Absolutely shocked. And if you believe that, I got some oceanfront property in Arizona. I'll sell you too.
On Samsung they got rid of a perfectly good screenshot tool and replaced it with one that has AI, it's slower, clunky, and not as good, I just want them to revert it. If I wanted AI I'd download an app.
AI SLOP!
Anyone who has been paying attention has been waiting for this enormous bag of shit to explode already.
AI is useless for most people because it does not solve any problems for day to day people. The most common use is to make their emails sound less angry and frustrated.
AI is useful for tech people, makes reading documentation or learning anything new a million times better. And when the AI does get something wrong, you'll know eventually because what you learned from the AI won't work in real life, which is part of the normal learning process anyways.
It is great as a custom tutor, but other than that it really doesn't make anything of substance by itself.
maybe if it was able to do anything useful (like tell me where specific settings that I can't remember the name of but know what they do are on my phone) people would consider them slightly helpful. But instead of making targeted models that know device specific information the companies insist on making generic models that do almost nothing well.
If the model was properly integrated into the assistant AND the assistant properly integrated into the phone AND the assistant had competent scripting abilities (looking at you Google, filth that broke scripts relying on recursion) then it would probably be helpful for smart home management by being able to correctly answer "are there lights on in rooms I'm not?" and respond with something like "yes, there are 3 lights on. Do you want me to turn them off". But it seems that the companies want their products to fail. Heck if the assistant could even do a simple on device task like "take a one minute video and send it to friend A" or "strobe the flashlight at 70 BPM" or "does epubfile_on_device mention the cheeto in office" or even just know how itis being ran (Gemini when ran from the Google assistant doesn't).
edit: I suppose it might be useful to waste someone else's time.
Artificial Incompetence
Honestly I can't say I've ever had a reason to use it on my phone.
It actually gets in my way every time it does something so that I have stop what I'm doing to kill it. Would love to be able to uninstall it
I found AI tools awesome for removing objects in photos or transcribing a conversation. Other than that it's useless because it's not reliable.
Ai sucks and is a waste of humanity’s resources. I hate how everything goes on buzzwords industry trends. This shit needs to stop and just focus on simplicity and reliability. We need to stop trying to sell new things every cycle
As an android user (Pixel), I've only ever opened AI by accident. My work PC is a mac and it force-reenables apple intelligence after every update. I dutifully go into settings and disable that shit. While summarizing things is something AI can be good at, I generally want to actually read the detail of work communications since, as a software engineer, detail is a teeeny bit important.
I hate it 🤷 I keep it turned off anywhere that I can.
Count me in!
