Yes but the party doesn’t start until Apple arrivesSeems like Apple is late to the party...
Yes but the party doesn’t start until Apple arrivesSeems like Apple is late to the party...
Salakhutdinov told FT that one reason for Apple's slow AI rollout was the tendency of language models to provide incorrect or problematic answers: "I think they are just being a little bit more cautious because they can't release something they can't fully control," he said.
I could check it out as well, just curious...Hmmm, interesting. I‘m from Zurich and a former Apple employee. If anyone knows more details of a whereabout or so, let me know, would be interesting to sneak a location and send some pictures😎
Hiring specialized technology experts to work in a secretive lab in the mountains of Switzerland? This is some James Bond villain-type behavior.created a "secretive European laboratory" in Zurich
You have missed that many people here talk like bots and reply to stuff in a very stereotypical way regardless of the news itself, e.g.:On what planet can Apple ignore AI as a fundamental core technology?
Good grief. Generally Apple is always late to the party. They sit back and watch the market then improve on the crap that is put out their "first". It is one of the reasons Apple is the best at what they do.Seems like Apple is late to the party...
OpenAI is not much small. One of the founders was Elon Musk. Also Microsoft has invested lots of money into it. Tho I agree that company is still comparably small in terms of investments to what Google/Alphabet hasgot beaten by a much smaller company called OpenAI
171b in cash and 110b and growing annual free cash flow gives me the faithGoogle has all the best AI researchers in the world and they invented the transformer behind the LLM revolution but it was the people at OpenAI who were the ones that knew what do to with the technology. Just because Apple has these same talented researchers doesn't mean anything will come out of it. Remember, Apple had all the money and talent in the world but they didn't know how to make a car. It takes people like Sam Altman and Elon Musk to make products from new technology that consumers actually want to use.
Also this is a friendly reminder that Apple couldn't add a functioning autocorrect to the iPhone keyboard until 2023. Autocorrect is more or less a more primitive version of an LLM. Let's not mention Siri because that's a dead horse. What makes people have any faith that Apple has any shot at all at being competitive in the AI race?
You have missed that many people here talk like bots and reply to stuff in a very stereotypical way regardless of the news itself, e.g.:
- Google adds whatever random feature in one of their apps
"Did you know that Google sells your data? I know this because I am very informed and smart"
- Apple does something for the environment
"Did you know that even after this Apple will still produce emissions? I know this because I am very informed and smart"
PS: Google does not sell your data to anyone. Google keeps your data for themselves. What they sell is a promise that the ad you will give to them will appear to the right person. But good luck explaining this to the average, "informed" but technically clueless guy.
QuickType suggestion in my language are not bad, but also not great. The vocabulary is rather limited in my experience. I think this is an area were Apple could still improve a lot. German has many very long words. A predictive algorithm that would be able to predict words after only a couple of characters of input would be VERY useful.Also this is a friendly reminder that Apple couldn't add a functioning autocorrect to the iPhone keyboard until 2023. Autocorrect is more or less a more primitive version of an LLM.
Actually Apple is doing exactly the opposite of what you’re fearing : they’re having the A.I run locally on your devices precisely to avoid the privacy issues. The fact that they hired some techs from Google doesn’t mean they‘ll want to have the philosophy as Google, just like hiring developpers from Microsoft doesn’t mean Apple wants to license MacOs to everyone.I'm still sorely disappointed with Apple.
I chose Apple to avoid utilizing Google (I'm slowly De-Googlizing, or whatever), and having its privacy-infringing technology baked into i/macOS would be the ultimate joke...
I don't mind APple purchasing start-ups instead of building something in house, but Google...
Come on...
THANK YOU! I keep hearing this arguement that “apple is late” and if anything apple has been here the longest. I think the confusion is coming from the terminology. Apple generally referred to it has ML all these years (machine learning) they rarely if ever called it AI. They’ve been ahead of the game in terms of image processing and other ML based tasks. They had the best hardware for it in mobile devices and frankly since AS they’ve been right there in computers. We will know soon enough if Qualcomm has caught up/beat them here soon.Who are Apple's competitors? Who is Apple trying to catch up with? These hires have been going on for years. AI/ML isn't some new focus for Apple. The rapid rise in popularity of LLMs might have caught Apple leadership off guard, but Apple is not new to machine learning. I'm not talking about Siri. Apple does tons of work with "AI" in images as well as text. It’s a major behind the scenes feature of iPhones and Macs.
If Apple is behind, why did it include a Neural Engine in iPhones starting in 2017? This was used for specific tasks like Face ID, but Apple didn't stop there. Apple has been putting a lot of effort into boosting up its Neural Engine. Apple brought the Neural Engine to the Mac with the M1 where it is used for many "AI" tasks. It is really good at running machine learning tasks, including LLMs. I can run local LLMs faster on my M1 Pro than I can on my 5900X with an RTX 3070. That is because of several factors, including the Neural Engine. That it can hold its own or beat CUDA on a decent GPU is impressive.
Apple doesn’t just throw things against the wall to see what sticks. There’s a longer plan to what Apple does. We are also in the very early days of LLMs. I love them more than the next guy, but they are not sustainably scalable. Something has to improve in their efficiency. That’s part of what Apple is working on.
I searched for some text on iOS Photos the other day, and was surprised (and impressed) to see that it was able to read and correctly identify the word in a photo of my scrawled cursive handwriting. I can barely read it at times, so hats off to Apple there.THANK YOU! I keep hearing this arguement that “apple is late” and if anything apple has been here the longest. I think the confusion is coming from the terminology. Apple generally referred to it has ML all these years (machine learning) they rarely if ever called it AI. They’ve been ahead of the game in terms of image processing and other ML based tasks. They had the best hardware for it in mobile devices and frankly since AS they’ve been right there in computers. We will know soon enough if Qualcomm has caught up/beat them here soon.
Apple just wants their own AI (iAI) that thinks fine woven products are the best thing since sliced gluten free bread.
This is actually a good point. Considering the obvious biases and pre-programmed responses we've seen from OpenAI and Google, I have no doubt that Apple's will be programmed to toe the Apple line.
I might agree if we were talking about hardware. But this is different. IMO, Appe do not have the best record when talking about services. For example, iCloud lags behind Google and Microsoft. And other services such as Apple Music, Apple TV+ and Apple Arcade are behind compared to the competition. And looks like they are also behind in generative AI. That could be the reason there are rumors of Apple is in conversations with Google and OpenAI.Yes but the party doesn’t start until Apple arrives