Apple Continues Losing AI Experts to Meta

[MNWildFan]

If this isn't a sign of them throwing in the towel on Apple Intelligence then I don't know what is

They were behind the 8 ball right away with overpromising but underdelivering, so at this point, they may as well just keep what they have with AI right now and just be ok with that and for anything else continute to build on just reaching out to other models like ChatGPT, Gemini, Grok, Sonnet, etc etc

 

[visasavo]

im glad even fanboys know how ****** apples AI efforts are

 

[neuropsychguy]

This is an important point. You still have to verify the solutions the coding agents spit out. And I don't think that's something that will change soon, even with more resources. A LLM will never "know" if the solution it spits out is correct. It might be, if it has "seen" the problem in its training data. But what about original and new problems?

That's why I verify. How I use them makes it difficult for the code to be wrong. I know what I want and what the code should do. It's simply using existing packages and software (for work I do that's not coding statistical analyses) so as long as the input (my data) is accurate, the output will be accurate. I still like to verify.

About your question: "But what about original and new problems?"

I had this come up recently with some research. I needed to process data and used an LLM to help me create a novel pipeline to do that processing of my research data. It used existing tools but I put them together in a way that no one had done before. Without an LLM I would have done the same thing, it just would have taken me longer to code. There was verification of the outputs based on specific hypotheses. I'm being vague to keep anonymity but I knew the code was correct because the end result showed something that we'll just say was biologically accurate.

The great thing about these tools for me is they save me dozens to hundreds of hours in a year and allow me to do things I would not otherwise have the time to do. My own verification mainly comes from doing similar work for years without the benefit of these tools so I know what I want, how to do it, and what the result should be. LLMs just help me go through that process faster and produce better and better documented code than I'd produce myself.

 

[JulianL]

I know that articles about Apple's AI woes are a great opportunity for snarky comments, and I agree that Apple's offerings so far are well behind the curve with every new news story making the future of AI (or at least conversational AI) at Apple look bleaker and bleaker, but my primary feeling about all of this is one of sadness.

I started out with Google Assistant on Google/Nest smart speakers and then moved on to Alexa on Amazon devices. I've also played around with Homepod (1st gen) and Homepod Mini devices but have never bought any because of how rubbish Siri is (in my opinion).

Neither the Google/Nest nor the Amazon devices & associated software have completely satisfied me - mostly hardware deficiencies for Google/Nest and software deficiencies for Amazon Alexa - so I'd love to see Apple dig itself out of this hole and come out with something that leapfrogs the competition and delivers the smart speaker experience that I want. If Apple could do that I'd happily swap out all my smart speakers and displays for Apple devices and I'd also have that same capable AI assistant built into my watch and phone as well.

Of course these AI experts aren't just evaporating, they are going to other companies so more capable AI devices will be released by someone, but I already have Apple wearables and I trust Apple more than most other companies (I'm looking at you Google) to release quality hardware for things like smart speakers and to then not seemingly abandon (no product refreshes for ages) whole product lines a few years later (looking at you again Google) but sadly every news story I read about Apple's AI efforts seems to make the chances of my hopes for Apple AI turning into reality a more and more distant prospect.

 

[Vulkan]

Machine Learning is the new Computer Engineering. People are going to chase the moolah and go where their time is best compensated. Everytime Apple "loses" an engineer to Meta, its not a setback. It opens the door for a new engineer to find its place at Apple and begin anew.

 

[vantelimus]

But how is all of that going to benefit me? because the creators of AI keep telling everyone how it is going to revolutionize the world. When the word 'world' is used in ANY AI documentation, that means me, a member of the public. If AI is not benefiting the public right now then it is very disingenuous of AI creators to say it will help everyone/help the world. Is it going to help me get to work quicker? Is it going to help me cook my meals quicker? is it going to do all my shopping for me and get it sent to me safely? is it going to help me be more productive at work?

If it's use is currently limited then AI creators should restrict there optimism on where and how it will be used.

Reread the list of things. It isn't just AI creators who are benefiting.

For you personally there's lots to come. Here are things I can think of off the top of my head: Cures for diseases. Gene therapy. Personalized and better quality healthcare. Personal assistants that will help you with complex tasks and education. Automated household management. Improved and safer transportation -- and not just self driving cars, but across every mode of transportation from cars to rail to air travel and space travel. Enhanced financial management that will help you not only with prosaic tasks like budgeting, but investing, fraud detection and monitoring, financial and tax planning. Real-time language translation of speech and text. Better and customized and customizable entertainment.

And this does not mention all the things that might come from new AI-aided discoveries in math and the sciences that result in products we cannot imagine today.

So, yeah, improved transportation systems and better traffic control will help you get to work quicker and safer. It certainly is on track to do your shopping for you. It already is used to alert you to things you might want to buy. It already helps me cook more quickly and more diversely. Instead of searching cookbooks for that recipe I made two years ago to refresh my memory, ChatGPT gives me everything I need to know in one prompt. It will make you more productive at work. If you are in any kind of information/white-collar work and you aren't using it already, you are falling behind.

 

[turbineseaplane]

Apple AI has already done the most useful thing it can - making 16GB base Macs. Now it can go.

I just wish Apple Intelligence also "required" more SSD space (to up all the base storage levels)

 

[digimc]

Others may disagree but I never really felt that Apples heart was in it when it comes to AI.

I think Apple just thought they would throw some features together hoping that not too many users would care or expect to much.

That’s why I think Apple was caught off guard when the crap hit the fan over and the advanced Siri features they promised but could not deliver. I think Apple thought that they could advertise these cool features and make a few commercials and most people would just forget about it for a while giving them time to work on a slow roll out.

I will just never believe that Apple thought AI would take off like it did and as fast as it did.

I think the proof is in the pudding. If Apple was aggressively trying to advance AI they would be the ones going out there and poaching talent. I really never see Apple being a leader or innovator in the field of AI.

 

[turbineseaplane]

I think the proof is in the pudding. If Apple was aggressively trying to advance AI they would be the ones going out there and poaching talent. I really never see Apple being a leader or innovator in the field of AI.

As an aside, what is Apple really leading and innovating on that isn't an extension of what they were already doing?

(Just a reminder that Apple Silicon has essentially been a thing since the A4 chip in the iPhone 4 that released 14 years ago. It was already well under development while Steve Jobs was still around, somewhat incredibly.)

 

[ThailandToo]

Dude, even Tim Cook isn't getting paid what Meta is paying these people. Apple may underpay, but what Meta is doing is stupid and unsustainable.

Also, let's not forget the tens of thousands of tech workers these companies have laid off over the last 5-or-so years.

If these people can snatch up these multi-multi-million dollar positions, they should do it, but don't expect it to last long. The bubble pop is coming.

It will burst in time but the next five to ten years will be a boom. Expertise in AI is critical for these tech companies. Can’t just rest on Steve Jobs successes and hope to milk the existing products and grow services. Must innovate again.

 

[vantelimus]

The one thing I know AI will not be used for and this for things that make companies and their owners extremely wealthy. The post about DeepBlue winning the noble peace for work on proteins got me thinking. Will AI be used to find a cure for cancer, one of the most profitable areas of medicine there is. Look at all the other medical problems and disabilities that have no cure but yet make those trying to find a cure extremely rich.

Will AI be used to make buildings stronger and more energy efficient. Will AI be used to solve energy problems, like how to make it more cheaply. AI could solve much of the worlds problems but by doing so it would make those who profit from world problems go from very rich to being poor. I seriously doubt AI is going to be used towards the things we feel it should be used towards because far to many rich men and women would be very scared they would be about to lose their wealth.

It is AlphaFold2, made by DeepMind (owned by Alphabet), whose inventors won the Nobel Prize in Chemistry for protein folding.

The answer to your other questions can be summed up as: AI will be used to enhance any product or service that where knowledge and reasoning are applicable. For example, AI won't necessarily find a cure for cancer, but it will be used by researchers and likely play a significant role in finding cures for different cancers that might not otherwise be found without AI due to the complexity and scale of information that must be processed.

 

[laptech]

It is AlphaFold2, made by DeepMind (owned by Alphabet), whose inventors who won the Nobel Prize in Chemistry for protein folding.

The answer to your other questions can be summed up as: AI will be used to enhance any product or service that where knowledge and reasoning are applicable. For example, AI won't necessarily find a cure for cancer, but it will be used by researchers and likely play a significant role in finding cures for different cancers that might not otherwise be found without AI due to the complexity and scale of information that must be processed.

Thank you for point out my error.

As for the second part, I disagree with the notion of complexity because it's AI, nothing is complex for AI. Well that is what AI creators are trying to sell to the public and businesses, that AI can solve anything.

 

[vantelimus]

As long as the investors win, the Board will allow Crook to keep ruining the company.

The BOD defines what the company is supposed to do. Your opinion on whether the current strategy is ruining the company, for better or worse, is irrelevant as long as the BOD is happy.

 

[turbineseaplane]

It will burst in time but the next five to ten years will be a boom. Expertise in AI is critical for these tech companies. Can’t just rest on Steve Jobs successes and hope to milk the existing products and grow services. Must innovate again.

I wish they'd innovate a small phone and and iPad Mini Pro

There are two purchases where my money remains on the sideline and off their books.

 

[clg82]

Apple has never been about elaborate answers to search questions. Apple is a device company with tightly integrated hardware and software that maximizes the user experience. Once AI has matured, become more reliable, and become commoditized, Apple can be expected to partner with an AI provider like they did Google search to provide such results.

Well at least we can both agree than that Siri was an abysmal failure, and that Apple won't be able to innovate their own AI solution like they claimed for the latest iPhone, and they'll have to go through a 3rd party to achieve that.

 

[cameltahoe]

Well at least we can both agree than that Siri was an abysmal failure, and that Apple won't be able to innovate their own AI solution like they claimed for the latest iPhone, and they'll have to go through a 3rd party to achieve that.

Siri, Alexa and others were not and are not a language model. They are a voice activated front end. Apple has the Foundation model for local language model use. They are different things.

 

[AeroEd]

I've said it before and I'll say it again: Meta is being dumb here. Soon AI development is going to become commonplace and everyone will know how to do it. Also, Meta could have the best AI available but people won't use it because Meta is horrible. This is the biggest waste of money ever.

How many billions use what’s app, insta, facebook etc 🤣

 

[iPhoneFan5349]

If meta wants to offer these people in excess of $100m EACH, Apple would be foolish to even try to counter it. Let meta do a lot of the sloppy early work, and then when it becomes a commodity (which it will) just contract for it on the cheap. In the meantime, just partner with Perplexity or someone.

Yes, because apple needs the money, right?

 

[Ghengis LeMond]

Hope it won't happen. Claude is too good to be ruined by Cupertino crowd.

What would a Claude/Siri offspring look like?

 

[Amazing Iceman]

Interesting that Meta is not mainly poaching from nVidia, Alphabet, MS, etc., but from Apple.
I wonder what they are looking for, what is it that they want to copy…

 

[Hewhowalksbehindtherows]

A fourth Apple artificial intelligence expert has left the company to join Meta, reports Bloomberg. Bowen Zhang, who was on Apple's foundation models team, is the latest employee to abandon Apple for Meta.

The leader of Apple's foundation models group, Ruoming Pang, was one of the first Apple AI researchers to join Meta. Since then, several employees who worked under him have also left for Meta. Meta is aggressively hiring for its Superintelligence Labs, an AI division that's building advanced AI systems capable of performing at or beyond human-level intelligence.

Meta CEO Mark Zuckerberg has been offering AI engineers massive compensation packages to lure them from other companies, and Pang reportedly received over $200 million. Pay from Meta reportedly includes a high base salary, a signing bonus, and stock awards, and the money offered to Pang exceeds the compensation of almost all Apple employees except for executives. Presumably, the other AI engineers that left Apple have also received offers that Apple isn't willing to match.

Last month, OpenAI CEO Sam Altman said that Meta had been offering signing bonuses as high as $100 million. Meta has hired engineers and AI experts from Apple, OpenAI, and Anthropic. Bloomberg says Apple is "marginally increasing" the pay of its foundation models team, but is not paying at the level that Meta is.

With Apple losing key employees to Meta, it could continue to struggle to catch up in the AI race. Competitors like Google and Samsung have much more advanced AI features already, and this year, Apple was forced to delay promised Apple Intelligence Siri features until 2026.

Apple has been restructuring its AI teams, with AI efforts now overseen by Apple software chief Craig Federighi and Mike Rockwell, who led Apple Vision Pro development. Rumors suggest that Apple is considering using technology from Anthropic or OpenAI for future AI features, including an LLM version of Siri, rather than its own models.

Apple's discussions to rely on third-party AI technology have reportedly led to falling morale on the foundation models team that is now losing employees to Meta. Multiple engineers are reportedly actively interviewing for jobs at other AI companies, while Apple executives are aiming to reassure team members that it remains committed to in-house AI development.

Article Link: Apple Continues Losing AI Experts to Meta

This is a bunch of garbage-trash-talk. "With Apple losing key employees to Meta, it could continue to struggle to catch up in the AI race." Could it be that you don't know much?

 

[Nuno Lopes]

I see space for a disruptive OS coming with AI, natural voice input and eye tracking baked in its “kernel”.

It may take some time, 10 years maybe. The thing is, I don’t see Apple or Microsoft coming up with it.

Meta, Google and others have more to gain with the push.

… and it will probably start on a phone.

 

[Loud_Texan]

META, it's a trojan horse! Its a trap!

 

[dba415]

Makes sense. Even if the money were comparable between Apple and Meta, it still is preferable for top AI talent to jump off the sinking ship that is Apple AI. There's no better way to tank your career than to stick with Apple Intelligence.

 

[cjsuk]

Removed this. Duplicate.