In this article, Mike Thornton offers his predictions for what Apple might be planning for 2024 in terms of new hardware, peripherals and their use of AI.

So far, Apple has released M3 powered models of the MacBook Pro in 3 models and two screen sizes as well as the long overdue refresh of the 24-inch iMac. What of the other models?

M3 MacBook Air

We expect Apple to release new M3-powered MacBook Air models in the first quarter of 2024, probably around March. We have learned that the new models are codenamed J613 and J615, although it is not expected that they will bring any new features to the table other than an M3 processor. Mark Gurman from Bloomberg said in his December 10th 2023 edition of his Power On newsletter…

“I expect the company to market this new model toward owners of the M1 MacBook Air or Intel Macs rather than people who already have an M2 machine.”

As the 15-inch MacBook Air was released in June 2023, we don’t expect Apple to release an M3-powered version until the third quarter of 2024.

M3 Mac mini

As the Mac mini was only updated in January 2023, this makes complete sense. Based on their track record, Apple will want at least 12 months out of the M2 models before considering an upgrade.  So, we expect M3-powered Mac mini desktop computers to be released sometime in the first half of 2024.

M3 Mac Studio

The M2-powered Mac Studio desktop computer was announced at WWDC in June 2023. This machine has hit the sweet spot for pro-audio users and many other users, especially Mac Pro users. Even before the M2 Mac Pro and M2 Mac Studio were announced in February 2023, we asked Apple Mac Pro 2023 - Is It Already Dead In The Water?

In June 2023, Apple announced both a Mac Studio with M2 Pro and M2 Max options as well as an M2 Mac Pro. As a result, in our article Which to Buy - Apple Mac Studio or Mac Pro 2023? we reviewed the Mac Studio and Mac Pro alongside each other.

So when do we expect a new Mac Studio? We could see an M3-powered Mac Studio as early as the third quarter of 2024, but we don’t expect an M3-powered Mac Studio until very late in 2024.

M3 Mac Pro

The last model in the Apple Mac range is the Mac Pro and the Mac Pro is a similar story to the Mac Studio as Apple released the M2-powered models together in June 2023. In the 13th August 2023 edition of his Power On newsletter, Bloomberg’s Mark Gurman dropped a bit of a bombshell with regard to which Mac computers would benefit from the M3 Ultra SoC. He said…

“Macs that will use the chip: Mac Studio and Mac Pro (if Apple continues making those).”

This concern may have a good foundation. For example, Apple Insider’s Malcolm Owen has said…

“There's a chance that Apple's fumbled Mac Pro update could lead to it giving up on the model for the moment.

With the introduction of the Apple Silicon model, some elements of the Mac Pro's utility became less impressive, in part due to the existence of the Mac Studio. With both the Mac Pro and Mac Studio able to use Apple Silicon, and with the Mac Studio being generally cheaper to purchase, it made the Mac Pro a hard purchase for anyone just wanting high levels of performance.”

The Mac Pro’s claims of being ‘expandable’ took a serious dent with the Apple Silicon model. This latest variant does not support additional graphics cards at all, and the SoC architecture also means that the memory cannot be extended as it is an integral part of the SoC design.

The range of SSD storage options is the one area that is still expandable. If this interests you, do check out our article Upgradable SSD Is Back In A Mac!

Coming back to the PCIe slots, one area of good news, especially for those who want to use the PCIe slots for additional storage, is that six of the seven PCIe slots are 4th generation, which Apple claims provide…

“twice the bandwidth for audio and video I/O, networking, and storage cards”. Apple goes on to claim that “the latest PCIe gen 4 SSD cards, for example, are capable of a colossal 26GB/s”.

All in all, the future of the Mac Pro is by no means certain. Unfortunately, the SoC design has compromised most of the key features users wanted from the Mac Pro model. 

If Apple does decide to release an M3-powered Mac Pro, we don’t expect to see it until late 2024 at the earliest, and it may not be until 2025.

USB-C Magic Peripherals

Even though Apple has been using USB-C on its Apple Mac computers for some time now, all the Magic peripherals currently ship with USB-C to lightning adapters because the peripherals still have lightning connectors on them. In early December 20203, Bloomberg’s Mark Gurman wrote…

“The company is also preparing revamped versions of the Apple Pencil and Magic Keyboard accessories, which it will sell alongside the new iPad Pro. The new Pencil — codenamed B532 — will represent the third generation of the product. The company released a new low-end model in November. The new Magic Keyboards — codenamed R418 and R428 — will make the iPad Pro look more like a laptop and include a sturdier frame with aluminium.”

It is predicted Apple will release the updated range of Magic peripherals - Magic Mouse, Magic Trackpad, and Magic Keyboard with a new USB-C charging port, at the same time as it launches the new M3 MacBook Air in March 2024.

One other thought: wouldn’t it be great if Apple could move the Magic Mouse's port anywhere but on its bottom?

Artificial Intelligence

In September 2023, Bloomberg’s Mark Gurman wrote, “Apple sees 2024 as the year it plans to enter the generative AI race”, with reports that Apple will spend $1 billion per year on artificial intelligence.

It is especially interesting as, so far it has been reported that Apple seems to have missed the boat with regard to artificial intelligence developments when ChatGPT and other AI tools shook up the technology industry. In October 2023, Mark Gurman wrote in Bloomberg…

“The company has some catching up to do. Apple largely sat on the sidelines when OpenAI’s ChatGPT took off like a rocket last year. It watched as Google and Microsoft Corp. rolled out generative AI versions of their search engines, which spit out convincingly human-like responses to users’ queries. Microsoft also updated its Windows apps with smarter assistants, and Amazon.com Inc. unveiled an AI-enhanced overhaul of Alexa. All the while, the only noteworthy AI release from Apple was an improved auto-correct system in iOS 17.

As I first reported in July, the company built its own large language model called Ajax and rolled out an internal chatbot dubbed “Apple GPT” to test out the functionality. The critical next step is determining if the technology is up to snuff with the competition and how Apple will actually apply it to its products.”

The plan so far is for Apple’s software engineering group to add AI to a future version of iOS, with the aim to fill it with features running on Apple’s own large language model, which is expected to improve how both Siri and the Messages app can handle questions and auto-complete sentences, mirroring recent changes to competing services.

Another part of the plan is for Apple’s software engineering teams to look into integrating generative AI into development tools like Xcode to help 3rd party software developers write new applications more quickly in a similar way to services including Microsoft’s GitHub Copilot that offers auto-complete suggestions to developers while they write code.

Apple is also looking into ways that generative AI can be used to help users write in their applications like Pages and to be able to auto-create slide decks in Keynote. These are not unlike tools that Microsoft has already launched for its Word and PowerPoint applications.

We understand that Apple is starting to test generative AI for internal customer service apps within its AppleCare group.

There are also discussions starting within Apple as to whether it might be possible to deploy generative AI fully ‘on-device’ rather than as a cloud-based setup. The advantage is that an on-device approach would be faster and help safeguard privacy but make it harder for Apple to update its technology and adapt to a fast-changing environment. Whereas having their large language models in the cloud would enable more advanced operations.

Or is there another way? Apple has published a new research paper titled "LLM in a flash: Efficient Large Language Model Inference with Limited Memory." It shows Apple has developed solutions to technical AI issues that other companies seem to be ignoring. Apple plans to use massive large language modules on lower-memory devices like the iPhone.

In the paper Apple says that it…

"tackles the challenge of efficiently running LLMs that exceed the available DRAM capacity by storing the model parameters on flash memory but bringing them on demand to DRAM."

This means that although the whole LLM still needs to be stored on-device, it can be accessed using flash memory as a kind of virtual memory, which is similar to how the macOS works for memory intensive tasks. The Apple research paper goes on…

"Within this flash memory-informed framework, we introduce two principal techniques. First, 'windowing' strategically reduces data transfer by reusing previously activated neurons... and second, 'row-column bundling,' tailored to the sequential data access strengths of flash memory, increases the size of data chunks read from flash memory.

The practical outcomes of our research are noteworthy, we have demonstrated the ability to run LLMs up to twice the size of available DRAM, achieving an acceleration in inference speed by 4-5x compared to traditional loading methods in CPU, and 20-25x in GPU.

This breakthrough is particularly crucial for deploying advanced LLMs in resource-limited environments, thereby expanding their applicability and accessibility."

This means that LLMs (large language models) can be used on devices with limited memory and Apple can implement AI features in more ways across more devices.

Unlike most of what Apple does, which it keeps under wraps as much as possible, Apple is choosing to make some of their AI research public. Firstly, it published another Al related research paper, proposing an AI system called HUGS, which generates digital avatars of humans and now this second paper on using flash memory to host LLMs.

This is interesting, especially in the context of of a report from Morgan Stanley, which concludes that Apple is ready to profit as consumers adopt AI-based tools in everyday life.

In their report Morgan Stanley says…

“Apple is not only further ahead in the field than it publicly states, but that as early as 2024, it will become one of six "key beneficiaries... [in] a catalyst year for 'Edge AI'."

Edge AI is Morgan Stanley's description for what happens when AI stops being a ChatGPT-like app and becomes embedded in a device’s operating system…

"[As] AI permeates into new consumer use cases, we expect the edge to become an emerging enabler of AI inferencing in 2024 given the benefits of lower query costs, improved latency, greater personalization, better data security/privacy, and easier accessibility.

While some investors question Apple's AI intentions, we believe Apple will be an AI enabler. We believe Apple will emerge as one of the key winners — or 'Edge AI Enablers' — in this race given the unique data from Apple's 2 billion+ devices and 1.2 billion+ users, Apple's focus on data privacy, and Apple's leading hardware, software, silicon and services vertical integration.

AI is traditionally deeply integrated across Apple's product and services ecosystem (think FaceID, TrueDepth Camera, Fall Detection, Smart Dictation, etc.). Other ‘Edge AI’ use cases that are likely to emerge on Apple devices include more efficient code development, integrating AI into Apple's native productivity apps, and AI-powered health, financial, and fitness assistants, amongst others."

But that is not all. Morgan Stanley believes that Siri will become "the ultimate smart assistant."

"[Using] LLMs [AI Large Language Models] to power Siri as the ultimate virtual assistant is, to us, the clearest 'killer AI app' Apple can deliver to market, and reports indicate this integration could come to Apple devices as soon as next year [2024].

We believe Siri powered by LLMs could become an 'intelligent layer' that sits in-between a users iPhone and the websites/applications the access, becoming an all- important gateway to many of the brands we use today."

Morgan Stanley expects that WWDC 2024 will be a key turning point of what Apple has to say about its AI development. Morgan Stanley, says that it will be…

"closely watching for AI-related commentary during future product launches, and the rollout of AI-enabled services (and/or developer tools), with Apple's next WWDC in June 2024 a likely key catalyst."

Apart from having the capital to invest in AI, Morgan Stanley believes that Apple has three advantages over its competitors…

"We believe Apple will be a Gen AI enabler given 3 distinct advantages, unique data from an installed base of over 2B diverse edge devices and 1.2B users, unmatched privacy standards, and vertical integration of hardware, software, silicon and services that enables a more powerful device to process Edge AI workloads."

What Are You Looking Forward To From Apple?

You have read our predictions of what we expect Apple to release in 2024 and why. What would you like to see Apple release in 2024? Do share your thoughts and hopes of what you would like from Apple this year in the comments below.