On Sunday, Elon Musk, one of the richest men in the world, escalated his fight for control over artificial intelligence by disclosing the source code for his version of a chatbot.
A creation of xAI, the business Mr. Musk created last year, Grok is meant to respond to questions with a tongue-in-cheek tone reminiscent of the science fiction book “The Hitchhiker’s Guide to the Galaxy.” Despite being separate from X, xAI’s technology has been included in the social media network and is taught using user postings. Those with access to X’s premium features can inquire about Grok and get answers.
Through a practice known as “open sourcing,” which allows anybody to access and use the code, Elon Musk entered a contentious discussion within the artificial intelligence community about whether or not this makes the technology safer overall.
Although he hasn’t updated it since, Mr. Musk, a self-described supporter of open source, did the same thing with X’s recommendation system last year.
Although there is still work to be done, Mr. Musk wrote on Sunday in response to a comment about open-sourcing X’s recommendation algorithm, “This platform is already by far the most transparent & truth-seeking (not a high bar, I know).”
The switch to open-source chatbot technology is the most recent exchange of blows between Mr. Musk and OpenAI, the company that created ChatGPT and was recently sued by the volatile entrepreneur for violating its pledge to follow suit. After leaving OpenAI a few years after its founding, Mr. Musk made the case that Microsoft, Google, and other digital behemoths like them shouldn’t have complete control over such a significant technology. Microsoft is a close collaborator of OpenAI.
According to OpenAI, it will try to have the lawsuit dismissed.
Since the technology’s rise in popularity last year, there has been much debate about whether or not to make generative artificial intelligence (A.I.) open source. This technology can produce realistic images and videos as well as human-like text responses. The question of whether the coding that powers artificial intelligence should be made public is a contentious one in Silicon Valley. While some engineers contend that the technology is too powerful to be left unchecked, others maintain that there are more advantages to openness than disadvantages.
Mr. Musk solidified his position in the latter group by disclosing his A.I. code; this move may allow him to outpace rivals who have advanced the technology more quickly.
When the code is made public, other businesses and independent software developers will be able to use and adapt it to create their own chatbots and other artificial intelligence systems. Facebook and Instagram’s parent company, Meta, has also made its LLaMA artificial intelligence technology publicly available. Open sourcing has also been used by Google and Mistral, a well-known French start-up.
As the CEO of Tesla and the owner of X and SpaceX, Mr. Musk established xAI last year with the goal of helping people “understand reality.” He stated in November that a quarter of xAI would be owned by investors in his $44 billion take-private agreement for X.
Mr. Musk has declared that chatbots should be able to handle any topic, branding as “woke” businesses that control their technology to steer clear of controversy.
In a statement published on Friday, Mr. Musk stated, “If an AI is programmed to push for diversity at all costs, as Google Gemini was, then it will do whatever it can to cause that outcome, potentially even killing people.”
Nonetheless, there is a strong commercial component to at least some of the rhetoric around open source. With the most potent and possibly most well-liked chatbot on the market, OpenAI leads the competition and has no incentive to make its code publicly available.
On the other side, Mr. Musk and xAI are attempting to catch up and may help level the playing field by making their code open source and encouraging others to further the technology.
Arizona State University computer science professor Subbarao Kambhampati has maintained that the safest course of action for current A.I. technology is to make it open source. However, he went on to say that for that reason, businesses like Meta and xAI weren’t necessarily making the technology open-source.
The main artificial intelligence scientists at Meta, Elon Musk, and Yann LeCun, he argued, “are not the best messengers for this argument.”
