Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Why Apple believes it’s an AI leader–and why it says critics have it all wrong

arstechnica.com

Post reply on HN
Print view View original

101–110 of 217 posts

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#101

Post by throwaway202020 »

moandcompany wrote: post #100
throwaway202020 wrote: post #81

You can't be AI leader when every AI leader is staying away from you by 10ft pole. When Ian Goodfellow joined Apple, there was literally rain of criticism on him from ending his career as researcher to bowing down to money. I don't know of any researcher who does want to have continued research career willing to join Apple. They simply don't allow that kind of freedom or publishing results. While Apple has some stron…

I'll argue that AI leadership is not limited to just research, and an under appreciated aspect of leadership is making actual things useful for people (e.g. applying AI/ML technology to problems, and making those solutions common or ubiquitous).

It would be hard to be leader without having leaders in the field at your disposal. There are literally 20,000 papers getting published each year. While stunning gains are being reported every month, it would be very hard for you to identify good implementable papers with state of the art results unless you are active in the field. Many of the techniques are amalgamation of other dozen techniques. So its even more harder to add/subtract from a given paper to productize the results. Keeping up with these papers and techniques is literally a full time job which is why you need researchers at your disposal if you are working on these problems. Unfortunately, researchers are almost never willing to just give up research and become your next engineering dude. They are in the field because they love research, freedom and can publish. So the viable way is to have a research lab which you can tap into on demand and have options for rotations. Many people believed that this can be done without having your own research lab but that doesn't often work out. If you approach a researcher in academia for consulting, you have to count on them being free from academic duty, grad students and other commitments. They will typically do 80/20. If you have your own research lab, you can turn this around to 20/80.

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#102

Post by throwaway202020 »

jointpdf wrote: post #34

Setting aside all the other work Apple is doing in this area, can someone explain why iMessage autocorrect is unacceptably bad (to put it diplomatically) in the era of weapons-grade language models? Its behavior is totally bizarre. It’s like an underpowered chess engine that makes flagrant blunders: capitalizing random words in the middle of a sentence (rock -> Rock), the context sensitivity of an actual rock, forcin…

Many teams at Apple are pure junk. When Steve Jobs basically fired iCloud folks in public, lot of people thought he went too far but little they knew how bad that team really was and still is. These folks are now center of the Apple universe but can't build a simple app that can efficiently sync the photos on Windows. Other teams are iTunes (one of the worst widely deployed Windows app ever), Apple Music (just look at the obscene UX - in 2020), App Store (good luck deleting installed app if you have too many of them - in 2020) and of course auto-correct (fails to fix spellings in 2020 that MS Word 2000 used to fix!). Surprisingly these folks have kept up with mediocracy regardless of otherwise stringent culture at Apple.

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#103

Post by TheOtherHobbes »

jolux wrote: post #19
skybrian wrote: post #18

Earlier quoted context omitted.

It's not my area, but I read Google's AI blog and they seem to be doing machine learning research that ends up in its phones, such as various camera improvements and offline Google Translate. It seems like it's the sort of thing you do when you can, but often it comes as a second phase after getting it to work in the data center. I didn't see anything in this article that was obviously unique to Apple.

I think a lot of people have underrated the benefits Apple reaps in this area from being so far ahead in chip design and controlling their hardware. As regards to successfully productizing ML I’ve recently become convinced that Apple has been more successful at it than Google so far. Translation was already well within Google’s wheelhouse and they had been doing it for years, and their work with cameras is going to i…

Apple is bad at AI because Apple is bad at software, and increasingly bad at common sense. Being good at hardware won't compensate for that.

Example: Apple Maps regularly thinks I need help finding my way home from locations close to where I live. Some basic practical intelligence would understand that I have visited these places before and there's a very good chance I already know my way home.

It would know that I would appreciate a route if I'm a couple hundred miles from home at a location I've never been to. But a shopping trip to the next town fifteen minutes away? Thanks, but no - that's Clippy-level "help."

IMO the company is stuck in the past, its software pipeline is so badly broken the quality of the basic OS products is visibly declining, and it's unprepared for the next generation of integrated and invisible AI applications.

Siri was a magnificent start but it seems Apple not only failed to build on it or take it further but actively scaled it back to save a few dollars in licensing fees.

Google is doing better by looking at specific applications - like machine translation. But because it's an ad company with a bit of a research culture it can't imagine an integrated customer-centric general AI platform - which is where the next big disruption will come from.

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#104

Post by thdrdt »

l33tman wrote: post #99
jointpdf wrote: post #34

Setting aside all the other work Apple is doing in this area, can someone explain why iMessage autocorrect is unacceptably bad (to put it diplomatically) in the era of weapons-grade language models? Its behavior is totally bizarre. It’s like an underpowered chess engine that makes flagrant blunders: capitalizing random words in the middle of a sentence (rock -> Rock), the context sensitivity of an actual rock, forcin…

To be fair, the android keyboards (like Gboard, googles own) are also bad. It's a tough problem for sure, but they haven't gotten better for the last 6-7 years at least, in my experience, and that is the point of the parent post I guess.

Is this true? Lately I noticed a change. Autocorrect now suggests based on previous words. I am not sure this use to be the case.

For example if I type "this is not the" it suggests the next word might be "case".

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#105

Post by tomduncalf »

jointpdf wrote: post #34

Setting aside all the other work Apple is doing in this area, can someone explain why iMessage autocorrect is unacceptably bad (to put it diplomatically) in the era of weapons-grade language models? Its behavior is totally bizarre. It’s like an underpowered chess engine that makes flagrant blunders: capitalizing random words in the middle of a sentence (rock -> Rock), the context sensitivity of an actual rock, forcin…

Same here, it’s painful and actually seems to get worse the more I use my phone. I’m pretty sure it capitalises words found in your contacts - fine for surnames, very annoying for Home, Bank etc.

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#106

Post by fauigerzigerk »

godelski wrote: post #40
jointpdf wrote: post #34

Setting aside all the other work Apple is doing in this area, can someone explain why iMessage autocorrect is unacceptably bad (to put it diplomatically) in the era of weapons-grade language models? Its behavior is totally bizarre. It’s like an underpowered chess engine that makes flagrant blunders: capitalizing random words in the middle of a sentence (rock -> Rock), the context sensitivity of an actual rock, forcin…

This also happens to me on Android. Working in ML, this is something to point to when people hype it up. We've come a long way, but we still got a long way to go. As for specific errors, does rock capitalize when "the" is before it? (The Rock) The models learn from how users write. But when you think about it, it is actually fairly difficult to predict what words are going to be, you can test it with your friends. Ha…

Could autocorrect not wait until it sees more of the sentence?

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#107

Post by Abishek_Muthian »

lacker wrote: post #84

"Google is an amazing company, and there's some really great technologists working there," he said. "But fundamentally, their business model is different and they're not known for shipping consumer experiences that are used by hundreds of millions of people." What does he mean by this? Google search and Android are both used by more than a billion people. YouTube over 2 billion. These are all bigger than any Apple pr…

He's just being subtle and putting it short of Apple's usual marketing speak - "We chose not to sell Data" perhaps as a goodwill to his former company.

Btw,

>they're not known for shipping consumer experiences

IMO, difference is not in selling 'consumer experiences' but selling 'consumer aspirations', which you rightly pointed out is not for many.

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#108

Post by shrimpx »

throwaway202020 wrote: post #81

You can't be AI leader when every AI leader is staying away from you by 10ft pole. When Ian Goodfellow joined Apple, there was literally rain of criticism on him from ending his career as researcher to bowing down to money. I don't know of any researcher who does want to have continued research career willing to join Apple. They simply don't allow that kind of freedom or publishing results. While Apple has some stron…

Apple has fallen way behind in software/apps, and maybe they’re ok with it. The hardware and OS are good, and some staple apps for creators are good. But at everything else they seem really lethargic or absent. For example, as a web developer Apple has had zero impact on my tooling and knowledge. But their laptop is nice and I have one, which may be good enough for them.

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#109

Post by posedge »

zwegner wrote: post #82
jointpdf wrote: post #34

Setting aside all the other work Apple is doing in this area, can someone explain why iMessage autocorrect is unacceptably bad (to put it diplomatically) in the era of weapons-grade language models? Its behavior is totally bizarre. It’s like an underpowered chess engine that makes flagrant blunders: capitalizing random words in the middle of a sentence (rock -> Rock), the context sensitivity of an actual rock, forcin…

I recently discovered a neat trick for fixing the duck problem, that is, adding swear words to the iOS dictionary: add a contact with the chosen words in the name fields. E.g. I have a contact named "shit shitty fuck fuckin fucking oh shit"

I find it both amusing and ridiculous that you have to go to such great lengths to be able to type a good swear word on the iphone.

Re: Why Apple believes it’s an AI leader–and why it says critics have it all wrong

#110

Post by rsa25519 »

jointpdf wrote: post #34

Setting aside all the other work Apple is doing in this area, can someone explain why iMessage autocorrect is unacceptably bad (to put it diplomatically) in the era of weapons-grade language models? Its behavior is totally bizarre. It’s like an underpowered chess engine that makes flagrant blunders: capitalizing random words in the middle of a sentence (rock -> Rock), the context sensitivity of an actual rock, forcin…

Ah, my favorite is autocorrect changing `its` to `it's`. Makes my grammatically correct sentences incorrect :-)
Post reply on HN