Nearly every IT organization has been funding generative AI for the past two decades, and Apple is rumored to be the first to release new AI-based tools for iPhone users immediately with iOS 18. However, maintaining such tools costs a lot, and Meta only confirmed that it will take decades for the business to turn a profit from AI.
CEO Mark Zuckerberg spoke to buyers about Meta AI, Meta’s fresh associate that has been integrated into apps like WhatsApp and Instagram, and has also been added to Ray-Ban Meta clever glasses as Meta shared the benefits of its first governmental quarter of 2024 on Wednesday.
The future of Meta AI seems to be in Zuckerberg’s hands, pretty excited. The executive did point out that it will take decades for the company’s investments in conceptual AI to be successful. He noted that “experiments are a very great lengthy-term investment” and that the “initial signs are quite good”. However, earning from relational AI is “going to take several years”. (via The Verge)
After Zuckerberg confirmed that the company may spend even more wealth on AI, Meta’s stocks dropped more than 10% on Thursday. By the end of the year, Meta made a recent announcement that it intends to purchase about 350,000 Nvidia H100 AI bits. Each of these cards costs between $20,000 and $40,000, which means that Meta will spend billions of dollars.
The costs associated with conceptual AI are, in fact, very great because it requires a lot of processing power to train and use such devices, as well as the server system. Platforms like OpenAI’s ChatGPT cost users $20 to let them get more sophisticated tools like GPT-4 and DALL-E. Meta, on the other hand, doesn’t act out monetizing its AI through advertisements in the future.
What about Apple’s conceptual AI?
With all of this in mind, how does Apple compare to its conceptual AI? We won’t have definite information until WWDC 2024 in June, but based on rumors, we can speculate.
Apple has been exploring both on-device and server-based speech concepts, according to various publications. While Bloomberg’s Mark Gurman just reported that Apple’s novel AI features may be powered entirely by an online LLM, there are also studies that Apple is negotiating with Google to embrace Gemini on its websites.
If I had to guess, I would say that the business would, in fact, employ both offline and online models. Offline models have the drawback of not being as powerful and necessitate more advanced chips, but they can also lower Apple’s server-related costs.
Apple might offer free offline AI for some basic tasks while more advanced tools could be paid for online and be paid for, perhaps in combination with an iCloud+ or Apple One subscription. This could help the business pay for hosting costs or potential agreements with third parties like Google.
The company already puts features like iCloud Private Relay under a paywall, so charging for a “Siri+” wouldn’t be a surprise.
What about you? Would you pay to use Apple’s generative AI? In the comments section below, let us know.
