AI Race APPLE, NVIDIA, AND COMPETITORS BATTLING FOR INDUSTRY DOMINANCE

With Apple unveiling a new method for training large language models (LLMs), rumors circulating about new acquisitions at the Cupertino company, and chip announcements from NVIDIA, the AI race is heating up in what could be the biggest year ever for technology.

- 

APPLE’S RACE TO CATCH UP

Nothing has stood still for long in the world of technology, but when it comes to artificial intelligence (AI), Apple is struggling to compete with established and emerging players. We only have to look at the current state of Siri compared with the likes of OpenAI’s ChatGPT. But the tide could be turning, with Apple making significant strides with the development of a new training method for large language models (LLMs) that adeptly combines textual and visual data. This breakthrough, encapsulated in their research paper “MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training,” represents a pivotal advancement in AI technology, propelling the capabilities of AI systems to new heights.

The MM1 model leverages a rich dataset that includes image-caption pairs, interleaved image-text documents, and purely textual information. This diverse dataset is instrumental in training the AI to excel in tasks such as image captioning, visual question answering, and natural language inference. The model’s architecture and training methodology enable it to interpret and generate language grounded in a combination of visual and linguistic cues, a competency crucial for tasks requiring an in-depth understanding of the visual world. The MM1 model’s in-context learning prowess is particularly noteworthy, especially in its 30 billion parameter variant. This configuration demonstrates an exceptional ability for multi-step reasoning across multiple images and few-shot “chain-of-thought” prompting, facilitating complex problem-solving with minimal examples. This technology will likely be