Apple is reportedly still developing its generative AI technology, but it is already being used by employees behind the scenes.
Bloomberg News‘ Mark Gurman has revealed that Apple employees are using a ChatGPT-esque AI chatbot called “Apple GPT” to prototype future features, summarize text, and answer questions based on the data it has been trained on.
Gurman also said that Apple is exploring ways to integrate its LLM into other areas of operations, such as using it to assist customers with AppleCare.
Here are some possible use cases for Apple’s generative AI technology:
• Providing customer support: Apple could use its LLM to create a chatbot that could answer customer questions and provide support. This could free up human customer service representatives to handle more complex issues.
• Generating creative content: Apple could use its LLM to generate creative content, such as poems, stories, or scripts. This could be used in a variety of applications, such as marketing campaigns or product development.
• Improving product recommendations: Apple could use its LLM to improve its product recommendations. This could be done by providing personalized recommendations based on a user’s past purchases and interests.
• Creating new features: Apple could use its LLM to create new features for its products. This could be done by using the LLM to generate ideas for new features or to test out new features before they are released to the public.
Mark Gurman for Bloomberg News:
[Apple is] now all-in on large language models (LLMs) and is working on creating a generative AI product for as early as next year. Moreover, the company has already built an internal ChatGPT-like service for employees. It helps them prototype future features, summarize text and answer questions based on data it has been trained with.
The core of this LLM work is a new framework known as Ajax. It was used to create the ChatGPT-style app, dubbed “Apple GPT” by some inside the company, and it’s likely to have many more offshoots. The LLM project is a cross-company effort spanning the software engineering, machine learning and cloud engineering organizations.
Apple hasn’t yet determined what its eventual generative AI product will be for consumers, but there a few areas where you can bet the company is looking. For one, it’s easy to imagine a major Siri upgrade to make the assistant more conversational and add the ability to handle more tasks on behalf of the user. Imagine asking Siri to write an email about an upcoming meeting rather than having to actually dictate the message.
Beyond consumer products and the internal “Apple GPT” tool, the company is already looking to push generative AI to more parts of its operations. That includes using the technology to help its AppleCare support staff assist customers. And I’m sure that’s only the beginning.
MacDailyNews Note: In May, Apple posted a number of job listings seeking generative AI talent.
Please help support MacDailyNews. Click or tap here to support our independent tech blog. Thank you!
Support MacDailyNews at no extra cost to you by using this link to shop at Amazon.
Unfortunately, Apple GPT being used internally won’t generate any revenue for Apple.
First place “AppleGPT” will be used publicly is handling support requests. Then handling Siri requests. It’s unlikely that Apple will sell ChatGPT-like subscriptions, but rather will use “AppleGPT” to keep labor costs down.
If Apple turns “AppleGPT” into a revenue center that’s because some engineer imagines some brand-new app based on “AppleGPT” that generates tons of money in a way I can’t can’t currently conceive of.
Ajax is new?
Apps generate revenue for both Apple and developers. So, quite obviously, generative AI for SwiftUI code is a money maker.
I’m looking forward to this. The current “chat” bots are good for the mundane stuff (such as “make this class comply with the Codable protocol”), but a built-in Swift-specific tool would be fantastic.