Elon Musk has officially unveiled his open-source artificial intelligence (AI) model.
“We are releasing the base model weights and network architecture of Grok-1, our large language model,” Musk’s xAI said on its website Sunday (March 17). “Grok-1 is a 314 billion parameter Mixture-of-Experts model trained from scratch by xAI.”
The billionaire had announced last week that he was going to roll out an open-source version of Grok, the AI chatbot introduced in 2023.
This followed a lawsuit by Musk against artificial intelligence giant OpenAI — a firm he helped found — accusing the company of departing from its mission to come up with AI that helps humanity, in favor of seeking profits.
Musk was among the founders of OpenAI at its birth nine years ago, but stepped back from the company’s board in 2018.
“To this day, OpenAI, Inc.’s website continues to profess that its charter is to ensure that AGI ‘benefits all of humanity,’” his lawsuit said. “In reality, however, OpenAI, Inc. has been transformed into a closed-source de facto subsidiary of the largest technology company in the world: Microsoft.”
OpenAI said earlier this month it will ask a judge to throw out Musk’s suit, writing on its blog that it remains committed to a mission to “ensure AGI [artificial general intelligence] benefits all of humanity.”
In that post, OpenAI said that it realized building AGI will require “far more resources than we’d originally imagined,” and that the company and Musk agreed in 2017 that a for-profit entity would be necessary to obtain those resources.
Meanwhile, some legal experts say Musk could have an uphill battle. In an interview with PYMNTS earlier this month, attorney Seth Kugler, of the Cupertino, California-based firm Grellas Shah LLP, argued it would be difficult to show in court that OpenAI jettisoned its original mission because it licensed GPT-4 and kept elements proprietary.
“In order to prove that this is a breach, Musk must show that GPT-4 is artificial general intelligence,” said Kugler, who focuses on AI and corporate governance.
“AGI is an artificial intelligence that can broadly perform any task as well or better than a human,” he added. “Accordingly, it can teach itself to do new tasks outside of its original training — basically a full substitute for high-level human intelligence.”
For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.
We’re always on the lookout for opportunities to partner with innovators and disruptors.
Learn More