We may earn compensation from reviewed products. Learn about our editorial policies.
For a world so fixated on AI, Apple has made a point of avoiding publicly saying the phrase. The biggest tech company in the world seemingly doesn’t want a slice of the AI pie. Until now…we think. The unofficially named Apple GPT is on the horizon.
Rumours have been circulating about Apple’s development of its own generative AI tools, with a particular focus on an AI chatbot like ChatGPT. Read on to find out what Apple GPT actually is, its potential release date, and Apple’s generative AI future.
What is Apple GPT?
Apple GPT is a generative AI project that Apple is reportedly working on. Sounds conclusive right? So far, there has been no official confirmation from Apple, but there are some firm rumours.
GPT stands for ‘Generative Pretrained Transformer’. This refers to a type of AI model that uses deep language learning to generate human-like text. OpenAI’s ChatGPT is the most notorious example, but there are others like Google Gemini.
Apple’s version is internally referred to as Apple GPT and will be an AI chatbot similar to those already on offer.
According to Mark Gurman at Bloomberg, Apple has created its own framework called “Ajax” which they are using to build large language models (LLMs). LLMs are what inform the natural-sounding responses given by generative chatbots. For example, Google’s LLM used for Bard is called PaLM 2.
Apple Ajax is expected to be more advanced than models like GPT 3.5 which was recently superseded by GPT 4. GPT 3.5 was trained on 175 billion parameters whereas Apple’s generative AI model reportedly sits around 200 billion parameters.
This is slightly odd for Apple. Most of the time, they want to be the best everywhere they go. Just look at Vision Pro as an example. But GPT 4, the most advanced AI model around, was trained on 1.76 trillion parameters. That makes a huge difference.
This suggests that Apple is holding back, or making multiple smaller AI models trained for different purposes. All we can do for now is wait for more concrete information.
Release Date
There is no information at the moment regarding the release of Apple GPT or some other form of Apple AI.
However, Apple’s public avoidance of the subject doesn’t necessarily mean a formulated AI strategy doesn’t exist. Only a fool would believe that Apple is winging it.
Analysts like Ming-Chi Kuo predict that Apple GPT won’t be released in 2024. However, there could be an announcement at its annual Worldwide Developers Conference (WWDC) confirming Apple GPT exists. Something similar to Vision Pro happened in 2023; announced at WWDC, not released until 2024.
This leaves 2025 as the most likely year we will see some Apple generative AI services. More than anything, this shows how far behind Apple is on AI…or how little they care.
Features
Although details about Apple GPT’s specific features are limited, we can speculate on potential capabilities based on existing generative AI models and Apple’s areas of focus.
As with most tech companies, Apple already uses AI and machine learning in various products and services. These include enhancing photos in the iPhone’s camera app, powering the Spotlight search function, and providing autocorrect suggestions.
The main area that would benefit from better AI at Apple is Siri. If we’re being honest, Siri is pretty unhelpful and non-existent in most Apple user’s experiences. When was the last time you actually used Siri for a specific task?
As such, Siri has lagged behind rivals like Alexa for a while now.
With a more sophisticated generative AI model, Siri will become more conversational and capable of handling complex tasks. For example, integration with the Messages app would allow users to ask intricate questions and receive comprehensive answers.
Overall, Apple isn’t, and doesn’t want to be, primarily an AI company. Therefore, we can safely say that rather than being transformational to how smartphones and other products are used, Apple’s AI will likely make lots of small improvements to the user experience and efficiency gains.
Conclusion
Apple finally seems to be getting serious about AI after years of silence. However, there is no indication of a firm release date or any defining features.
Apple’s entry into AI will be a big event in the tech world, but not as intimidating as it could have been if they’d got involved earlier. Competitors like OpenAI are already streets ahead in terms of the complexity of their AI models.
Either way, Apple never does things lightly. They will want to excel at making AI a genuinely useful part of their products which is an enticing prospect. With that in mind, all we can do is wait for WWDC24 for any updates.
Apple Ring: Apple Smart Ring: All Updates, Information And Rumours
Apple GPT FAQ
Does Apple have a GPT?
A – Apple GPT is the informal name given to Apple’s generative AI chatbot project. It is being developed using Apple’s own framework called ‘Ajax’ which will also be the basis of Apple’s Large Language Model.
When will Apple GPT be released?
A – While there is no confirmed release date for Apple GPT, analysts speculate that Apple may make an AI-related announcement in 2024, possibly at WWDC. A release is most likely in 2025.
What features can we expect from Apple GPT?
A – Apple GPT’s specific features are not yet known. However, it is likely to enhance Apple’s existing AI-powered services, such as Siri, Apple Music, and customer support apps. Most of the features will make the user experience more efficient.
How does Apple GPT compare to other generative AI models?
A – Apple GPT is expected to leverage Apple’s expertise and resources to create a competitive generative AI model. However, Apple is believed to be behind its competitors in terms of generative AI development.
Does Apple have an AI strategy?
A – The short answer is yes, but it isn’t as developed as competitors like OpenAI or Google. Apple hasn’t made any AI plans public but they are developing their own generative AI chatbot.
