Apple has significantly increased its spending on conversational artificial intelligence (AI), investing millions of dollars per day, according to a recent report. While John Giannandrea, Apple’s AI chief, previously expressed skepticism towards AI chatbots, the company formed a dedicated conversational AI team four years ago to explore the potential of this technology.
Apple’s “Foundational Models” team, consisting of a small group of 16 individuals, focuses on conversational AI. However, the majority of the significant investment is allocated to training language models, which requires expensive hardware. Notably, OpenAI’s CEO, Sam Altman, revealed that more than $100 million was invested in the development of GPT-4.
In response to the growing competition in the chatbot sector, Apple has been working diligently on its own “Apple GPT” artificial intelligence project. Seeking to rival OpenAI’s ChatGPT, this project has become a major priority for Apple in recent months, as the demand for chatbot services and AI features in apps continues to rise.
Important to note is that Apple’s Ajax platform relies on Google’s Jax machine learning framework, operating on Google Cloud. Though Apple reportedly experimented with OpenAI’s technology, it ultimately chose not to enter into a contractual agreement with the company.
Insiders reveal that Apple is actively developing its own “Ajax” framework to support large language models. Presently, a select group of Apple employees has access to an undisclosed chatbot, referred to by some engineers as “Apple GPT.” However, access is restricted, and its applications are limited to internal use only.
Apple has not yet established a definitive strategy for incorporating AI-related products into its customer offerings. The existing Apple chatbot can answer questions using the data it has been trained with, but further advancements and broader application remain in the works.
Acknowledging the company’s relative lag in AI-based technologies compared to its peers, Apple’s primary AI objective is to develop a voice assistant akin to Siri. This assistant would be designed to automate tasks that involve multiple steps, potentially integrating multi-step voice-controlled automation in the upcoming iOS 18 operating system.
While Apple users can currently create multi-step actions on their devices using the Shortcuts app, the aim is to streamline this process and enhance user experience.
In a related development, Apple’s manufacturing partner Foxconn has been selected as the exclusive supplier of servers dedicated to training and testing Apple’s AI services. Foxconn, which already provides servers for Apple’s data centers, will construct these servers in a facility located in Vietnam. Additionally, Foxconn serves other leading companies in the AI sector, including Amazon, Nvidia, and OpenAI.
- Is Apple investing heavily in conversational AI?
Yes, recent reports indicate that Apple is spending millions of dollars daily on conversational AI.
- Does Apple have a dedicated team working on conversational AI?
Apple has a small team called “Foundational Models” focused on conversational AI.
- What is Apple’s goal in developing conversational AI?
Apple aims to create a voice assistant similar to Siri to automate tasks involving multiple steps.
- Is Apple competing with OpenAI?
Yes, Apple is developing its own AI project, “Apple GPT,” to rival OpenAI’s ChatGPT.
- Is Apple behind in AI technologies compared to other tech giants?
Apple has acknowledged being behind in AI-based technologies but is actively working to bridge the gap.
- Will Apple integrate AI automation into its operating systems?
There is a possibility that Apple may introduce multi-step voice-controlled automation in the forthcoming iOS 18.
- Who is Foxconn and what role does it play in Apple’s AI endeavors?
Foxconn is Apple’s manufacturing partner and exclusively supplies servers for training and testing Apple’s AI services. It also serves other prominent AI companies.