Elon Musk says the new large language model from his artificial intelligence lab xAI will be available to a select group from tomorrow. In a cryptic post on X, he claimed that in some respects it is "the best [model] that currently exists".
The world’s richest man launched xAI in July, bringing together a team of leading engineers and AI scientists to "understand the true nature of the universe" through AI. This came off the back of the growing success of OpenAI through its chatbot ChatGPT.
Musk was an early investor in OpenAI but left the board in 2018 after an apparent disagreement and a failed attempt to take over the company. He was also heavily investing in machine learning and AI through the self-driving vehicle division of Tesla at the time.
Why launch his own AI company?
The billionaire has been critical of many AI products, services and research efforts from big tech companies, claiming too much emphasis is placed on guardrails and controls. Earlier this year he said he wanted to launch a "maximum truth-seeking AI".
In contrast to these lofty ambitions, Musk has also warned that AI poses an existential threat to humanity and that in the future it will put an end to work and create a world where "no job is needed". It isn’t clear how much of that comes from his xAI research or is just opinion.
As well as Musk himself, xAI includes engineers who have previously worked for Google's DeepMind, Microsoft, OpenAI and Tesla. They have collectively worked on some of the most important technologies that power modern AI tools like ChatGPT and Bard.
What will the model be like?
The new xAI model is expected to be in direct competition with OpenAI's GPT-4, Google's PaLM 2 and Meta's Llama 2 with much of its training happening using Oracle's cloud network.
Musk has been in the U.K. this week at the AI Safety Summit. During a live conversation on X with British Prime Minister Rishi Sunak he declared AI has the potential to become the “most disruptive force in history.”
He said: “If you wish for a magic genie, that gives you any wish you want, and there’s no limit. You don’t have those three wish limits nonsense, it’s both good and bad. One of the challenges in the future will be how do we find meaning in life.”
During its initial announcement in July, Musk opined that for a "truth-seeking super-intelligence", it was safer to build it with curiosity and an interest in humanity as it will be more interested in keeping humanity safe.
That is part of what he hopes to achieve with xAI, a way to understand the universe and create a “good AI”. What format it will take, how it will be accessed by those given permission to use it or what the cost of the model will be have yet to be revealed.