Elon Musk has admitted that his move to develop his own AI company is to challenge OpenAI and Google, who have already rolled out AI-powered chatbots.
The entrepreneur said his motive is to “create a third option” in an industry already being dominated by OpenAI’s ChatGPT. Musk made this known in an interview with Fox News. He said he’s thinking of naming the company “TruthGPT” with the goal of creating a “maximum truth-seeking AI that tries to understand the nature of the universe” and that “hopefully does more good than harm.”
The Twitter CEO helped develop OpenAI, but exited the company in 2018 holding no stock. Amid the frenzy created by ChatGPT, which has triggered global AI race, Musk has been very critical of AI development, warning that it poses existential threat to humanity and could be the end of civilization if it is not regulated. He has also criticized OpenAI for deviating from the original plan to make the venture non-profit.
While Musk’s criticisms appear not to change anything, he has decided to create an AI version that will represent everything he had advocated. But he seems late to the party. Musk acknowledged that he is “starting late” which puts him in significant disadvantage to his rivals.
It is not clear how seriously Musk intends to pursue his plan. But the WSJ reported last week that the world’s second richest man is assembling researchers and engineers to start his AI project.
Musk is making the plans while working to transform Twitter, the microblogging app he acquired late last year. He had vowed to free the social media platform from censorship. He said on Monday that the US secret service had access to all Twitter accounts, including direct messages, under previous management.
Musk’s push to promote free speech on Twitter and his decision to develop TruthGPT, suggest there is a grand plan to make the social media platform part of his AI pursuit.
He said Twitter was working on a feature that would allow users to optionally encrypt their direct messages, and that it’s “hopefully coming out later in the month, but no later than next month.”
Musk has previously said that Twitter should offer Signal-style encryption, which is believed to be a way he intends to make conversation on the platform secure.
“If you are in a conversation you think is sensitive, you can just toggle encryption on and then no one at Twitter can see what you’re talking about,” he said.
But encryption may be in addition to his other plan to make Twitter a payment platform. Musk has merged Twitter with X Corp., a parent company he’d registered to bring his companies under one umbrella, in pursuit of his “everything app” called X.
ChatGPT Continues To Accelerate
Meanwhile, ChatGPT is not waiting to be caught as it now moves into the stock playground, LinkedIn News explains.
ChatGPT has begun to insert itself into the world of finance. Not only was the AI chatbot able to distinguish between “hawkish” or “dovish” statements by the Federal Reserve, it successfully determined whether headlines were good or bad for a particular stock, according to two academic papers. The research also shows that the technology developed by OpenAI can perform these and other tasks without being specifically trained to do so. This so-called zero-shot learning capability already exceeds earlier technologies, experts say. Chegg, an educational software maker, is combining quiz answers with GPT-4 to create a study aide called Cheggmate.
