US

Elon Musk has said he plans to create a “TruthGPT” to counter what he claims is liberal bias shown by a popular chatbot.

The billionaire Twitter, Tesla and SpaceX owner told Fox News his alternative to OpenAI’s ChatGPT would be a “maximum truth-seeking AI that tries to understand the nature of the universe”.

Musk also accused OpenAI, in which he was an early investor, of training its chatbot “to be politically correct”.

His commitment to building a similar service comes less than a month after he called for a pause in the training of powerful AI systems as they posed “profound risks to society”.

ChatGPT is a so-called large language model, which is trained on an enormous amount of data to interpret text and respond to prompts in a human-like manner.

A recent upgrade to the technology behind it, dubbed GPT-4, massively improved upon the chatbot’s ability to understand and complete tasks.

Users have demonstrated it writing poetry, solving computer bugs, and passing exams, while companies including Microsoft and Expedia have rolled the tech into their own products to help users with everything from searching the internet to booking holidays.

More on Artificial Intelligence

But there are concerns about the accuracy of the information shared by models like ChatGPT and Google’s rival service Bard, which have both been shown to occasionally spout outright falsehoods with supreme confidence.

Please use Chrome browser for a more accessible video player


2:16

Will this chatbot replace humans?

Musk backs regulation of AI

As the development of AI continues at a rapid rate, so do calls for regulation.

Musk said he was a “big fan” of governments taking action, saying the technology was “more dangerous” than the rockets and cars his companies are known for.

Self-driving cars made by Musk’s Tesla have come under increasing scrutiny, however, with more than 360,000 of the vehicles recalled earlier this year due to fears their software may cause a crash.

Click to subscribe to the Sky News Daily wherever you get your podcasts

Last week, the Chinese government published draft rules outlining how generative AI services would likely have to adhere to the same tight restrictions as the rest of the internet in China.

In March, Italy became the first country to outright ban ChatGPT while the country’s data protection authorities investigated its collection of user information.

Read more:
What is OpenAI’s GPT-4 and how does it improve upon ChatGPT?
AI generated newsreader debuts in Kuwait

EU law enforcement agency Europol also warned ChatGPT may be used by criminals and to spread disinformation.

And the White House is also inviting public feedback on how AI should be regulated in the US, having unveiled its “blueprint for an AI bill of rights” late last year to protect jobs and privacy.

When will we see ‘TruthGPT’?

Reports suggest Musk is still building a team for his AI start-up.

But he has already incorporated a new business called X.AI Corp, according to a Nevada business filing.

It was formed on 9 March and lists Musk as its director and his adviser, Jared Birchall, as secretary.

It comes after it emerged Musk had merged Twitter into a firm dubbed X Corp.