[mondobe]

I have a hard time seeing how this isn't obvious. 95% of everyday AI needs (for the people that even bother to interact with it) are covered by ChatGPT, and most of that is the same stuff that Google was handling before.

From personal experience, the only thing that changed when replacing the "old" Google Assistant with the Gemini-powered one on my Pixel was that it's no longer able to create reminders.

[ffsm8]

The only thing that changed for me was a that it couldn't control my smart Home devices anymore, nor activate navigation, nor send messages via Whatsapp (I.e. while driving).

Literally every thing I used it for got answered via "I cannot do that yet" after it randomly opted me into that. Pure garbage.

[piyush_soni]

Then why are you using it? I tried using Gemini once on my Pixel 6. Couldn't play music on Youtube music on verbal instructions, I switched back to Google Assistant. Will try it again after 6 months now. :)

[TeMPOraL]

Isn't that obvious? They tried to switch back to Google Assistant, but each time they asked Gemini, it said it can't do that yet!

[kyriakos]

My main use of Google assistant is to add things to my shopping list while I'm cooking. Switched to gemini, which I could have a human like conversation to get information about any topic but it just couldn't add items to the shopping list. I switched back. I didn't need a chatbot to keep me company I needed an assistant.

[ffsm8]

> Then why are you using it?

I'm not, I pretty much just accepted that Google doesn't care about usability whatsoever and haven't prompted it in a very long time.

To be clear, the only time I've ever used it was via "ok Google" in contexts in which I'm unable to interface with the phone directly, i.e while driving. If it doesn't work you'll learn that you can't start driving before queueing the navigation anymore. The voice assistant was a nice feature, but not important enough to waste my time trying to figure out which feature they opted me into and how to get back out of it.

[xur17]

In my case it just kinda.. switched over at some point, and frankly I didn't care enough to figure out how I might switch it back (if I even could). I had a similar frustration to GP that it stopped working for 100% of the queries I used to use it for.

That said, at some point it started working better, but there was a good 6-12 months where it was a tire fire.

[swatcoder]

I'm not sure that's true. Not everybody is hounding for information from the web in the first place.

Apple's approach of using current-boom AI to help you navigate and digest your own private trove of multimedia content (photos, videos, apps, notes, structured data, etc) is absolutely useful to people as well, and for some of us, one of the only personal uses of this AI that seems compelling at all.

I'm much more excited to have help finding that goofy picture of my cat by describing what I remember, so I can share it with a friend, than I am to have some chatbot dialog about entry-level Python with a hallucinating parlor trick.

But these features have to work, and work well, and work fast, and be widely known to work, before they'll really win the market. But that's going to take a minute and it might not even happen.

[itsoktocry]

>I'm much more excited to have help finding that goofy picture of my cat by describing what I remember, so I can share it with a friend, than I am to have some chatbot dialog about entry-level Python with a hallucinating parlor trick.

Hasn't Google been doing this forever? I can search random things in my photos (like pictures of an old car I owned).

[swatcoder]

For a bit, and to a degree, yes. Last-decade image recognition and tagging teased what might be possible and is genuinely useful.

The new LLM-ish tools promise that users can be more vague and casual in what language they use and more elaborate in how specific they mean to be; and that the queries (and operations) can span more diverse data sources.

[radlad]

Are there examples of new tools based on recent AI advancements that perform better than Google Photos image recognition?

[conradev]

Google highlighted the delta that recent advancements brought to their products:

https://blog.google/products/photos/ask-photos-google-io-202...

[gumby271]

Wow they announced this back in May and it's still not available for me.

[gsich]

Or better than Picasa almost 10 years ago.

[lancesells]

iOS has been doing this for a bit too. I don't use it enough to really know how good it is but I can definitely look for cats or people I know. Haven't used Apple Intelligence yet so maybe that's better as well?

[reportingsjr]

Google photos is way, way, way better than apple photos at this. It’s not even a competition.

I have my sister’s dogs named in my google photos library. Every time I a take a picture of either dog, they are automatically tagged and added to a shared album I set up for my sister.

I have nieces and nephews with photos from newborn age to 10+ years old, and it has managed to organize them across their growth and ages. It’s incredible. I can search for “<niece name> <vacation area>” and get every photo of her on a certain vacation to make a family scrap book.

Apple photos search and tagging is pitiful in comparison.

[WOTERMEON]

But one of the two you can have on your device and do not have to pay rent

[okasaki]

It's great for spooks too. Now they just have to exfiltrate the keywords describing the images instead of the images themselves.

[fao_]

I'm not sure that a hallucinated image of something is better than the original image when you're doing spywork.

The difference between whether someone has 4 or 5 fingers, whether they're holding a gun versus a random object, or whether they're mixed-race or caucasian, all seem like they would be pretty important things. Likewise, car number plates, signage in the photo that help identify where it is, metadata of the image itself (often more useful than the image), are all incredibly important. All of those are things that AI is absolutely terrible at lmao.

[qup]

He's not talking about generating images, he's talking about classifying existing ones.

[fao_]

> Now they just have to exfiltrate the keywords describing the images instead of the images themselves.

I'm not sure how a bunch of probably-correct keywords that miss a lot of the important aforementioned details in the image is more useful than the image itself, or it's metadata. Both of which would be lost. My point applies equally with respect to image classification, too.

[dylan604]

Yeah, it seems based on the advertising from the various AI vendors, they are showing its use by summarizing emails/phone call/etc. Things like being able to search text messages for info blah blah. The only one I've seen pushing online searches is Google, but that seems like duh! for them to be pushing. Circle something in an image and take me to a listing of that something for sale. Of course that's Google's direction.

But that whole find me something on my mutliple gigabytes of storage on my device account definitely seems like the mass appeal

[itsoktocry]

>But that whole find me something on my mutliple gigabytes of storage on my device account definitely seems like the mass appeal

This is such a mundane use of AI, but unsurprising Apple would sell it as revolutionary.

[cube2222]

Mundane or not, it’s actually useful, a meaningful improvement, and should work consistently well.

That is in contrast to a lot of fancy AI demos which are a great party trick, but fall apart in actual usage, with their reliability being “maybe it will work this time, maybe it won’t, just keep retrying :)”.

Apple is pretty well-recognized for usually being a bit late to the party, but at least delivering stuff that’s polished.

Just look at this thread of people sharing how Gemini broke all their commands and automations. The Apple Intelligence Siri on the other hand works just fine (even if new features are arriving slowly).

[consteval]

It's not very mundane, it's quite difficult to search for things when you don't know exactly what the thing is. Computers are laughably bad at this. I can tell a friend to find something in my room and give a vague description of the object. But with a file on my computer, I need to know actual content in the file. And images or video? Forget about it, if you don't remember part of the filename or what subdirectory it's in then it's gone.

[kredd]

The problem is, vast majority of smartphone usage is done for entertainment and social networking purposes (IG, TikTok, Twitter, HN, gaming, Netflix and etc.). If you’re mostly scrolling and consuming, I can’t imagine how current AI tooling can help you other than some summarization of texts. Sure, for productivity cases it might be legitimate, but that’s not what supermajority of people use a phone for.

[paul7986]

Im not sure people really know how to use an AI chat bot or understand its a more powerful, quicker and personally more fun way of getting information then just a Google search

Some personal examples of how I find it more useful (love to hear yours) and fun to use....

- Wanted to go on a hike an hour away from where both my friend (lives two hours west of me) and I live. Asked GPT what are some good hikes an hour drive away from both of us to meet & hike. With Google I have to do Many searches where GPT just provides the answer right away.

- I count calories and eat out everyday. GPT knows the calories of everything i eat as I eat at chains mostly (Cava, Panera, Starbucks, Chipolte). I tell it via voice what i just ate for my 1st meal, it calculates my calorie count and later I'll tell it what im having for my 2nd meal. It can also recall my calorie count from days ago. It does all this quickly vs. Google i'd have to do oodles of searches.

Usually Im using GPT the most when driving via voice and unlike Siri, GPT understands me and i can have whole conversations with it to get things done while driving.

[dragonwriter]

> Im not sure people really know how to use an AI chat bot or understand its a more powerful, quicker and personally more fun way of getting information then just a Google search

I can't speak to your personal entertainment experience, but AI chatbots are generally a slower, less accurate way of getting information than a google search. (Though Google polluting search results with a big, often inaccurate, AI result at the top narrows this a bit.)

[paul7986]

If you use it for research where you have to do many google searches vs. just ask one question like hiking question it's much quicker asking one question vs. multiple google searches to get ur answer.

[dragonwriter]

The thing is it isn't reliable enough to rely on the answer from just one question for anything that matters.

[williamcotton]

I don't really understand why it is acceptable to speak for others on this topic. It is fine if it doesn't work well for you. It is also fine if it works well for others.

These blanket statements lead to flame wars.

[reaperducer]

All of that sounds like the boring low-hanging life fruit that gets trotted out in videos by companies like Apple and Google as being "revolutionary." It's boring. It's staged. It's the easy stuff. It's well-off 20-somethings solving non-problems for other well-off 20-somethings.

Wake me up when I can say things like…

Hey, Google, are my custom license plates ready for pick up at the tax office?

Hey, Siri, ask my doctor to refill this medicine.

Hey, Alexa, how many charging stations are broken at the gas station on 16th street?

Hey, Google, why is this plant dying?

Hey, Siri, why are there so many people in my neighborhood today?

Hey, Alexa, did anything ever get done about that story in the newspaper from a couple of years ago about the Chinese slave labor being used to grow pot on illegal farms on the Navajo reservation?

"AI" just doesn't have access to the information required to do anything interesting or useful. And because so much of its information comes from the web, which is already so polluted on certain subject (gardening, travel) as to be useless, the AI becomes useless.

[hifromwork]

I see your point, but it's hilarious that I can ask AI "Write a python program that takes an URL parameter, connects over http to that URL, interprets the response as a CSV file and prints a sum of integers in the second column. Use argparse for command line parsing" - pure science fiction a few years ago - and you call this a "boring low-hanging fruit". Truly we humans get used to the good stuff quickly.

[reaperducer]

But still well within the described "well-off 20-somethings solving non-problems for other well-off 20-somethings."

[kredd]

I definitely agree, but again, if you look at the app usage times across every single demographic, those type of use cases are such a minuscule portion that it's just some noise. In my opinion, GPT/AI will cause more changes in workplaces, than in casual consumer spaces.

[mvdtnz]

No, the problem is that "AI" just isn't any good at almost anything.

[kredd]

You can argue that, but there are legitimate "AI" workflows that have already been adapted in workplaces where things function well enough. I'm sure 99% of it is just slop, most companies will fail, but some things will survive, become commoditized and taken for granted in about 5 years.

[rbanffy]

> If you’re mostly scrolling and consuming

Imagine an AI that popped up when you are reading something and warned you that information is false.

For instance, imagine something like https://theconversation.com/can-ai-talk-us-out-of-conspiracy... helping people discern about news and propaganda.

[bradyd]

Why would you trust AI, something that regularly makes stuff up, to be able to accurately determine that?

[notatoad]

the fun thing about the chatbots regularly making stuff up is that they almost always know when they're making stuff up. the hallucination problem isn't a problem of not knowing the facts, it's a problem of not knowing whether you want an accurate answer or a creative answer.

try asking chatGPT to only give you true and accurate answers and not make anything up.

[rbanffy]

I trust a defective AI a lot more than I trust Fox News ;-)

Now, more seriously, it'd need to put together a coherent argument and back it up with reputable sources, as just citing sources is very ineffective. The article I cited gives more details on possible approaches to that.

[jajko]

People eternally hoping that some new trick will finally make the other people understand how their side of the story/argument is one and only truly truthful. I guess this is old as mankind.

With your argument, the problem happens when given person goes to Fox news in the first place. Selection of the source has already been made, with its biases. Not much you can do or expect after this point.

Also, who curates the curator? Again an age old problem with no real, long term working solution in sight. No, you should not expect some statistical model to hold your hand through vast internet, while giving up any form of critical thinking, reasoning, or I guess any cerebral process altogether. Ultimate laziness. Since we know how much money there is in diet fad business, its safe to say this above will find its non-tiny desperate crowd.

[rbanffy]

> their side of the story/argument is one and only truly truthful.

There are no sides in objective reality. You might offer competing hypotheses and evidence for those, but we don’t need to do that to know that climate change is real, that it’s caused by humans, and that vaccines work.

[acdha]

That’d be useful but I’m pretty sure they’d get a massive backlash on Fox News and lawsuits filed alleging “being cancelled” within minutes of that shipping. It’s something we need but our current disinformation problem isn’t an accident but the result of decades of investment.

The less fraught one is warning users that they’re being scammed: https://arstechnica.com/gadgets/2024/05/google-wants-ai-to-l...

[rurp]

I can't imagine that working, even if that AI were more reliable than currently possible. Someone who already disbelieves credible reporting and objective sources is not going to be swayed by an AI telling them to disregard their fantastical conspiracy theories. Especially not in a near-future world where everyone is inundated with AI fakery.

That's even without considering the efforts that would be made to undermine any system that showed any effectiveness. A lot of misinformation, probably most of the stuff that gets traction, isn't random, it's serving a purpose and being pushed for that reason.

[ossobuco]

And who's going to teach the AI what's a conspiracy and what's not? Battles have been fought over what narrative certain Wikipedia pages should push; it's pretty hard to find information that can be objectively considered the truth.

[hifromwork]

That sounds like an useful tool for willing users, but if forced that sounds genuinely dystopian. And I'm afraid the people who would benefit the most (conspiracy theorists) wouldn't opt into that.

[rbanffy]

> And I'm afraid the people who would benefit the most (conspiracy theorists) wouldn't opt into that.

It's an unfortunate fact that the people most sick will refuse treatment until it's too late. Seen from the inside, insanity looks like lucidity.

[candiddevmike]

Changing to Gemini broke all of my smart home commands. Keeps trying to search/LLM how to turn off kitchen lights.

[lolinder]

I've heard horror stories and have held off so far in 'upgrading'. In the end I don't really want the fully flexible responses people are leaning into with these llm tools. All I want is to be able to give a precise instruction with my voice and have the machine reliably perform the action that it performed the last time I gave that instruction.

Since that seems to be an increasingly niche desire (at least as far as the product managers are concerned), I've been looking more and more seriously at setting up my own local voice assistant. My main barrier has been hardware—the mic arrays in the Home devices are surprisingly good and hard to beat with cheap off-the-shelf components, and you need a good mic for good STT.

[lancesells]

Yeah, I would like Siri to actually play the album I asked for and not something completely different phonetically from what I asked. Or even when I set an alarm and not be told "I can't connect to the internet right now" while I'm using my laptop connected to the internet. Or if my internet is down to actually use the speakers that I bought as speakers.

The hardware is really well done but the software is either over or under-engineered to a stupid degree.

[crustaceansoup]

It also sometimes asks to unlock my phone for commands that plain old Assistant was happy to do while locked. I haven't really found it useful at all yet, free ChatGPT is just better than free Gemini for "LLM stuff" and Google Assistant is better for "smart home stuff"

[connicpu]

I've been thinking about trying the OpenAI integration for home assistant[1], because controlling things in my home is primarily what I use my assistant shortcut for. The normal assistant works well enough but can be frustrating if you don't remember the exact phrasing it wants to activate a certain command.

[1]: https://www.home-assistant.io/integrations/openai_conversati...

[grahamj]

I have it set up with ollama. It’s… interesting. HA commands are provided to the model as tools so it works as well as the model is able to determine when and how to use tools. From experimenting with that and my own tool use code I’ve found that models vary greatly in their ability to wield tools and none that I’ve tried are exceptional.

It’s neat that you can intermix general chatting with HA commands but you’re probably going to find that the old assist is more reliable for commands. What I do like is that you can use a template as your system prompt so you can provide the state of a number of entities and then ask for them with natural language. That works well.

I have an Alexa/Echo voice announcement system set up and have recently tied that into assist so I can do automations like if the garage opens I prompt for “what is the state of the garage?” and announce the result. Makes it feel more humane than the same plain announcements all the time.

[TeMPOraL]

Do try it. I've been running it ever since it got integrated into the core, mostly to control A/C units around our flat, and it's the best voice assistant experience I had to date.

I mean honestly, how is it possible Amazon, Apple, Google and Microsoft[0] all keep screwing this up for over a decade now? I literally spent 15 minutes hooking up GPT-4 to the Home Assistant integration, and I was able to semi-reliably[1] control actual devices[2] like air conditioners and smart lights, in a completely natural and ad-hoc way, by talking to my smartwatch on the go, or to a phone, whatever was more convenient at the moment.

It's a really magical experience, a step closer to Star Trek reality. And what makes it possible is not just LLMs being able to deal with natural language, but more importantly, "bring your own API key" model allowing to cut away all the bullshit that FAANG assistants are stuck in.

--

[0] - Ever since they dropped MS Speech API in Windows, and did the Cortana thing. Some 15 years passed, at this point, and I'd still prefer to work with the Speech API than to touch any of the FAANGs' voice assistant - it worked, and worked off-line!

[1] - Works ~90% of the time; some 5% of the time the voice model (from Home Assistant Cloud) misunderstands me, and 5% of the time the LLM gets confused. It's still worth it, because I can actually talk to it like to a person, without thinking of style or grammar or magic keywords.

[2] - Which, given the level of integration of Home Assistant companion app with the phone, can be easily turned into an equivalent of on-phone voice assistant that can do more than the one I got from Google. Critically, there are ways to couple Home Assistant app and Tasker, so it's not hard to make it do arbitrary things on your phone. And, if you don't like low-ish code Tasker experience, you can trivially shell out from Tasker to Termux, at which point sky is the limit. Point being, an enthusiastic non-developer with minimal tech aptitude can beat Google and Apple at the voice assistant game today.

[add-sub-mul-div]

> Keeps trying to search/LLM how to turn off kitchen lights.

Teach a man to fish!

[rbanffy]

AI’s appeal depends a lot on the features. Battery is important for me (more than being anorexically thin), but I would love an AI that could screen my calls like a smart voice mail that asks questions.

Also, being able to talk to my mailbox asking questions about subjects mentioned in my e-mails would be a huge time saver.

Imagine a purely local Microsoft Recall-like thing that could answer questions about things you saw, or that read the news articles you went over quickly and answer complicated questions about them much later, at a time you just started to regret not having bookmarked it for future reading.

[est31]

Computational photography uses AI a lot (next to parametric approaches), and without these algorithms, smartphone cameras would be quite shitty. People do care about that. AI isn't just LLM chatbots.

[artwr]

I know that's true, but I find that the images on my Pixel are starting to have a bit of an eery feel, with some of the details looking more and more like AI generated images. I'd give back a bit of the quality for more "natural" looking images.

[wlesieutre]

That’s the line of thinking behind Halide’s recent “process zero” feature on iOS

https://www.theverge.com/2024/8/14/24220178/halide-camera-ap...

[renegade-otter]

Here you go: https://www.theverge.com/c/24208960/digicam-digital-camera-c...

[candiddevmike]

How do I disable that?

I'm having a horrifying realization that all of my pictures are "fake" in the sense that they don't match what I saw/experienced. Maybe it's time to get back into Polaroids.

[Terr_]

> I'm having a horrifying realization that all of my pictures are "fake" in the sense that they don't match what I saw/experienced.

I'd caution that judicious/proper post-processing is actually needed if you want that result, because of the differences between the sensors.

Your human experience comes from many small pictures taken by a set of lenses panning across multiple points in a scene with constantly adjusting exposure times and focal lengths, all biologically composited into what feels like a single moment.

Trying to fully replicate that with a single artificial picture is going to be deficient in certain ways.

---

Separately, a pet peeve of mine: Too many people have been subtly brainwashed into conflating the "like I was really there" with " like a Hollywood film camera was really there." Then the next thing you know your medieval fantasy game has lens flares in it for no good reason.

[izzydata]

I recently bought a real camera partially for this reason. I don't even mind how inconvenient it can be at times because honestly taking photos with a full camera rather than your phone is fun. While on the phone it has become quite dull in my opinion.

But knowing that the phone does a lot of software tweaking to get a picture to look similar to how good a full camera is made me want to switch. I think this was around the time that article about Samsung basically replacing a photo of the moon.

[jajko]

Get raw, and marvel at its imperfection and ugliness in many aspects. If given phone ain't giving up raw data, take any decent camera, raw sensor data has been part of it since its beginning.

But those pics will be probably further from perceived reality than those enhanced by software (lets not get retarded here and don't brand every data processing as 'AI'). Distortions not only of barrel type, waning brightness towards edges, moire, heavy vignetting, tons of noise, over/underexposition, maybe some dead pixels... thats not how I see my days go by.

[idle_zealot]

What does this mean? Sure, if generative AI is filling in features that weren't present, anyone would call that doctored or fake. But computational photography is mostly about recognizing patterns and filling in assumed detail... which is also how your visual cortex works.

[thrwaway1985882]

What makes a Polaroid any more "real" than an iPhone picture to you? Can any photo truly be real? (Deleuze has some interesting thoughts on the matter)

[ARandomerDude]

Right but in street surveys nobody knows that. Most people just call it "better camera."

[B1FF_PSUVM]

... then one day the tentacles come out ... https://i.redd.it/j1cr7pr6m7j71.jpg

[no_wizard]

Interpretation of this on its nose suggest the algorithms are the core feature not AI, as there is no artificial intelligence involved in these processes that I’m aware of.

If you actually peak under the hood they just pass through weighted selectors, no different than a switch statement

[dboreham]

That's what AI is. The weights are the clever part. Cameras have done this since the Nikon FA in 1983.

[dylan604]

what about the infamous recognizing the moon in the background that is just an over exposed white fuzzy circle and replacing with a stock image with full surface details? clearly there's some sort of ML/recognition of content within the image

[no_wizard]

Yes that would be ML as far as I’m aware, but the thing in reference is software compensation for image quality. Effectively smartphones automatically upscale photos by default. I think the only exception is when you choose to shoot in raw

[romwell]

>Computational photography uses AI a lot (next to parametric approaches), and without these algorithms, smartphone cameras would be quite shitty

Nah.

Smartphone cameras stopped being shitty a while ago, long before AI and computational photography hacks.

What you mean to say is they without AI, you'd know sooner that the smartphone maker put a cheap, shitty camera in your "premium" phone.

>People do care about that. AI isn't just LLM chatbots.

Yeah, it's also fake image generation featuring humans with a funny number of fingers.

What AI isn't is a camera.

[addaon]

> 95% of everyday AI needs

This assumes that the capabilities and use cases are unchanged. Yes, for the AI features available today, I suppose ChatGPT can do much of it -- I wouldn't know because it's not interesting or useful to me, so I don't use it.

But: If I'm deciding whether AI features are important to me in making a decision to spend money on a future phone, it's those future AI features that I will be assessing.

95% of my everyday needs for an external intelligence (besides my own) are covered by e-mail, text, and phone calls with other humans, with a trivial portion covered by nascent AI features. As this changes, and AI gets more capable of replacing human intelligence in these interactions (TBD if this happens in the next smartphone generation, or the next human generation, or further in the future), then I will /very much/ care that the electronic device that I use most often day-to-day has access to these capabilities, and will very much use access to those capabilities as part of deciding where to spend my money.

[njtransit]

You seriously assess "future AI features" when buying a phone? Have you heard the expression "a bird in the hand is worth two in the bush"? Also, what is the lifetime of a phone? How far in the future are these new features expected?

[addaon]

No, I assess current AI features when buying a phone, in the future. Any survey about what features customers value in phones they are considering buying inherently ask questions about future behavior, occurring in the future’s present.

[HumblyTossed]

Apple will just tell you you need an iPhone 17 because it has a more special "you're gonna love it" neural thingy onboard, and so your purchase of a 16 is null.

[grahamj]

> 95% of everyday AI needs (for the people that even bother to interact with it) are covered by ChatGPT

True, but it is an external service with all the privacy concerns that entails. I appreciate eg. Apple pushing local AI but at the same time I don’t think it needs to ship with the OS. Just provide an AI API apps can hook into then I can decide which models I want and where they run.

[jefftk]

> the only thing that changed when replacing the "old" Google Assistant with the Gemini-powered one on my Pixel was that it's no longer able to create reminders

Weird! Mine can still create reminders, as well as set timers and alarms.

[xvector]

"Everyday AI" hasn't even been built yet tbh. Where is my assistant that notices my flight has been delayed and reschedules my appointments and lets my contacts know?

[hn_throwaway_99]

I think your example highlights why this kind of "everyday AI" would be nearly impossible. If my phone automatically rescheduled my appointments and notified my contacts, without my explicit direction, I would be pissed as all hell. There are some "confirmation" notifications that could assist here ("We noticed your flight is delayed - would you like to reschedule these appointments?"), but even then, I'd say the majority of notifications I get these days are annoying and I spend a ton of time trying to figure out how to turn off annoying notifications while not totally silencing ones I depend upon. I'd have a difficult time believing that AI systems wouldn't just add to my list of over-burdensome notifications.

[michaelt]

If you were an important CEO or something, and you had a competent human Personal Assistant - would you allow them to reschedule your appointments if you were running late?

Seems to me the problems are (1) the "assistants" aren't anywhere near good enough to be trusted to make the right decisions, and (2) a trustworthy assistant isn't compatible with the adtech business model, so it's unlikely facebook or google would produce such a thing.

[shadowmanifold]

I think this is actually a deeper issue.

The CEO would trust the personal assistant to do this if they have a deep trust in the assistance competence. They would also need to know the assistant has a deep enough understanding of the their preferences to not do something they don't like. AI can mirror that.

More importantly though there will be consequences if the human assistant makes a big mistake and books the wrong flight. They would have to take responsibility for the mistake.

The LLM is always just going to write in text it is sorry if it makes a mistake. That is never going to be good enough for anything of consequence. The LLM would practically have to be omniscient in a way that is not going to be possible in a world filled with uncertainty.

So much of human activity is built around the network of trust that another human takes the blame if something goes wrong. So much activity involves coin flips and that someone takes the blame when the coin lands on heads but we bet on tails.

[no_wizard]

>If you were an important CEO or something, and you had a competent human Personal Assistant - would you allow them to reschedule your appointments if you were running late?

Yes, why wouldn’t I? I could also give them parameters like “if I’m more than 20 minutes late please re-schedule this” or “if my flight is delayed please let everyone know it’s delayed”

Why wouldn’t I do that? Presumably the person hired is competent to make determinations within parameters specified.

I could also let them know when it’s inappropriate to do this. Again, they should be competent enough to discern the differences between when it is and isn’t appropriate.

This could honestly be done by an algorithm if you give it the correct inputs and outputs and it could be fed updates, the only real limit is the fact that some of this isn’t exposed via an API either in a timely fashion or at all

[bluGill]

Important CEOs bring their trusted human assistant with them. Those of us at a lower level don't get that option, but at the CEO level you get to hire your own sectary and you spend years teaching them how you work and what to do in all the situations that come up and so they will make the right decisions often enough that you can trust them. Or they know they are not trusted to reschedule that appointment and so they don't.

That years of training is what we are missing. I don't think modern AIs can be trained in the way the assistants of old could be, at least not yet.

[no_wizard]

I don’t even think you need machine learning for this, it’s an API problem mostly like I said.

Being able to collate the requisite inputs from outside sources is the real problem. If you can’t do that reliably it’s simply hard to build an algorithm around it. Flights for example would require your calendar program to reliably pull data from an API regarding the flight information that is current and effectively real time. That’s the actual hard part, and this expands across services.

For all the advances we have made with computers and smartphones in particular they suck at meaningfully exposing a way to collate data sources and create actions around them reliably

[ziml77]

Personalizing any automation like that is a privacy nightmare. The system needs to know a lot about your preferences and the decisions you'd normally make yourself as well as your current circumstances since those will also influence your choices. How do you feed that into any AI without being problematic for privacy?

Having one running locally helps but it's still necessarily storing information that you might not want to have stored where someone could potentially retrieve it, either via some sort of exploit or by forcibly compelling you to give it up.

[Perz1val]

If I were a somebody worth a personal assistant/secretary, I'd be pissed about them doing such things without notice too

[bluGill]

If you had such a person they would know if you wanted that service of not and act accordingly.

[xvector]

Your problem isn't AI, it's shitty AI. Why wouldn't you trust an AI that was competent and understood what you wanted, like any good human personal assistant?

[vineyardmike]

I worked on a (failed) system that was supposed to help with this in the pre-ChatGPT era. The obvious limitations is getting the data, and scaling that to everyone. Learning about your flights, staying up to date on them, etc is such a daunting privacy-busting task that everyone is scared to start. Either you go use-case by use-case (flights, restaurants, calendars… etc) and never get traction or you just start scanning emails and open up huge data risks.

Today, this is nearly available, nearly. Probably only something Google/apple can realistically offer. Apple “intelligence” has started to read your notifications and rewrite them for you, so it shouldn’t be a big leap to listen for a United App notification and decide it’s urgent enough take action. Should be “trivial” for Google to do as well, and they could even run it server side to help without a phone present.

[literalAardvark]

That's agentic behaviour, and the first to offer it publicly are anthropic, who just opened a beta.

It's still pretty terrible though

[no_wizard]

I love how the goal post is being moved publicly as to what AI means.

AI is anything automated it seems, and now they’re being subcategorized into niches as to what they do, e.g. “Agentic AI”, “LLM backed AI systems” etc.

If it’s not real intelligence then it isn’t really AI, and I wish the world at large would call it out.

LLM, Machine Learning, Neural Networks etc are all great but none of them have true spontaneous intelligence or learning ability.

Please, someone point out how any of these systems have organic spontaneous learning ability for a subject it was not pre-data seeded on. This is a generally accepted measure of higher level sentience as far as I’m aware

[revscat]

> it’s not real intelligence

Hence the predicated “artificial”, and hence the downvotes you are currently receiving.

Your message is largely, if not entirely, a strawman.

[no_wizard]

None of these exhibit any accepted definitions of intelligence markers. The intelligence isn’t artificial, it doesn’t exist is my point. If you apply the commonly accepted definitions of what is considered display of intelligent behavior. One aspect of which is the ability to adapt to new circumstances that you haven’t experienced before.

There has been considerable success in programming computers to draw inferences, for example, but not actual reasoning. You can mimic some forms of reasoning but you can’t take one ML set - like recognizing photos with mountains, then expect it to correctly identify a similar geographical element - a hill. It can’t do that. It may correctly identify that it’s not a mountain but that isn’t the same thing as actually learning it’s similar to a mountain but not the same, which would be a rudimentary definition of a hill that an intelligent entity could conceivably use if it knew what a mountain was but not a hill.

Machine Learning was always a more honest place to have This discourse. I am indeed pushing back on the idea that we should be calling ChatGPT or anything like it intelligence.

It’s Machine Learning, clever algorithms, Large language Models, among other things, that are trained on ways to mimic certain aspects of intelligence, but it does not actually possess any real intelligence. Look at the LLM hallucination problem for example. It can’t be self corrected because it’s not an intelligent system.

Moving the goal post on what AI means (and pushing AGI as some new goalpost) is disingenuous, and relatively recent.

I’d care not if it wasn’t for the fact there is so much misinformation around capabilities and the future of AI, that it’s already negatively crept into policy making for example.

[xk_id]

I’m more annoyed by the sheer amount of effort and time we’re spending now as a society to rationalise the AI bullshit and argue against it. For example everything you said is valid, but you were provoked into painstakingly writing this elaborate rebuttal, by a bunch of parasitic VCs, who have zero intellectual integrity and are simply trying to advertise their investment.

[altdataseller]

1. How does it know how/when to reschedule that doctor's appointment that you need to call to reschedule? How do you know that receptionist won't hang up on your "AI agent" who tries to call them, b/c they think it's some sort of scam bot?

2. How does it know which contacts to contact? Does that acquaintance you talked to for some professional reason need to know your flight got rescheduled? What about that travel agency you talked to last night to confirm the flight?

[grahamj]

I would say the problem there is the requirement to phone a human to change a record in their database.

If they had an app then an AI assistant should be able to tie things together. Where things seem to be going is apps provide an intent-based API wrapper plus UI widgets to interact with it. That way assistants can operate them too.

[the_arun]

AI is usually about implementation details. Customers typically worry about solutions to problems. They don't care how solution is implemented - using AI or not.

[jeffbee]

I just asked my Pixel to remind me next Tuesday to pick the bananas, and it set that in Google Tasks. Which part of this doesn't work?

[larntz]

I have a Pixel also, and have issues with is since the change to Gemini. It works _most_ of the time, but every once in a while it'll tell me it can't set a reminder.

I use reminders often so I suppose it is a low failure rate.

But, when they first made the change to Gemini I had to switch back for a few weeks/months before it could set reminders properly.

[kspacewalk2]

I've set a reminder through Gemini, then tried to change its time. Told me there are no active reminders to change. Created another reminder with correct time. Got both reminders.

[lupire]

Failing to set reminders/alarms is core Google technology, #2 only to PageRank itself.

[jeffbee]

I said "Hey Google instead of reminding me about the bananas on Tuesday, remind me on Wednesday" and that worked. It's quite possible these things only work on Android 15 QPR 1, or only on a Pixel Pro, or who knows. Its capabilities have noticeably expanded in the last few months. One thing I know it still can't do, that the old Assistant did, is control a Sonos.

ETA: The joke is I don't even have Google Tasks installed, so I am not sure what effect those Tasks items would have. They might only surface on the side pane of Gmail.

[bluGill]

I want the reminder not on my device, but on my work computer which I'll be staring at when I need the reminder. Or in the bananas case I want bananas sent to our shared family grocery app (which isn't a google service) so that if my wife happens to be going by a store tomorrow she knows to stop and get some along with flour or whatever else we need to stock up on.

[blackoil]

I asked if to create task for Monday morning. It created one, Monday 12AM.

[Terr_]

Perhaps a step further: I do not need my phone to tell me an LLM-generated bedtime story involving a specified cast of characters, I just want it to reliably control existing functionality, dangit.

Like when a timed alarm is making the phone buzz in my pocket while I'm driving, I'm telling it to silence the alarm, and the response-voice is regretfully informing me that there are no alarms going off right now.