Elon Musk predicts that his company xAI could achieve general artificial intelligence (AGI) within the next few years, and possibly as soon as 2026, according to a new report from Business Insider. If it sounds like you’ve heard that one before, that’s probably because you have.
Musk predicted the same thing in 2024, saying that AGI will be achieved by 2025. Look at any calendar, and you will see that the end of 2025 is only a few weeks away.
“How long until AGI?” asked Logan Kilpatrick, head of product at Google AI Studio, in May 2024.
“Next year,” Musk replied, to which Kipatrick replied, “Big if it’s true.”
It wasn’t true, of course. But Musk has a long history, shall we say, and hope predictions about what his company will achieve in the future. And his predictions often have ulterior motives.
Remember when Musk made a lot of noise about the dangers of AI and worried that it could destroy the world? The billionaire signed a letter in March 2023 calling for a six-month moratorium on all AI development. It was revealed less than a month later that Musk was secretly building his own AI project on Twitter. In July 2023, Musk had officially announced the creation of xAI, the company that developed his Grok AI chatbot.
The CEO was not seriously concerned about the risks posed by AI. He was just frustrated that OpenAI was ahead at the time.
Musk’s treatment of AGI, or any new technology, depends largely on how well he can fire his companies at any given time. And the endless prospect of achieving AGI, whether you think it would be good or bad for the world, is helping to drive investment in AI technology, something that seems to be supporting the entire US economy at the moment.
A new report from Business Insider also claims that Musk told xAI employees that investments in the private equity firm are going well, with funding of “$20 billion to $30 billion a year.” An email to xAI with questions about the report was met with an automatic response that simply said “Legacy Media Lies.” Musk has a huge disdain for the media and previously had an automatic reply on Twitter that posted a poop emoji.
Part of the problem in discussing AGI is that there is no agreed upon definition. As IBM explains, we will have achieved AGI when artificial intelligence “can match or exceed human cognitive abilities in every task.” But obviously defining words like “cognitive skills” and “any activity” is very complicated.
Some people like to define AGI as a form of self-awareness that can make artificial intelligence more like humans. Instead of repeating words from its training data, the AI will understand itself as a form of consciousness. People in that camp are excited and/or worried about that point of view because they think it could be the start of a robotic revolution and AI’s attempt to destroy humanity. Musk has been vocal about those fears, though he’s backed off recently.
In the absence of massive robot armies, achieving AGI in the present day with a system that hates humanity would probably look like a 1970s sci-fi movie. Colossus: The Forbin Projectwhere non-humanitarian programs include nuclear weapons programs to threaten the world. We don’t really have advanced humanoid robots a Terminator 2 the situation at the moment.
But Musk is working on that too. He predicts that Tesla will produce one million humanoid Optimus robots a year within the next five years, and they’ll be watching over your kids. You just need to find a way to get Optimus to work without using a phone before all that happens.
Who knows? AGI could be magically achieved in the next few weeks, and perhaps Musk’s old prediction will come true. But he’s a millionaire and has another forecasting deadline coming up soon. Back in October, Musk told Joe Rogan that he would demonstrate a flying car later this year.
