Artificial intelligence has been on everyone's mind lately, including famed Shark Tank investor Mark Cuban.
Cuban took to Twitter May 18 to express his thoughts on the kind of evolutions generative AI -- like OpenAI's ChatGPT -- will soon experience.
DON'T MISS: Cathie Wood Says Elon Musk's Tesla Is Leading an Important Movement
"I'm a believer that LLMs need to evolve from being Language models to being Knowledge models (LKMs)," Cuban wrote. "To get there, they will need access to significant IP and data, the most important of which will not be free."
Cuban is curious which of the three big AI creators -- Google (GOOG), Meta (META), and OpenAI/Microsoft (MSFT) -- will be the first to pay for access to this data, and how much they'll be willing to shell out.
Cuban believes that the coming competitive AI environment will center around ownership over large quantities of data, something that will give billionaire inventor Elon Musk a significant leg up.
"All of this is why I think Elon Musk is starting his own @truthgpt, or whatever he will call it. He gets to take the entire Twitter firehose to train or feed any open source model and have a competitor to the big 3," Cuban wrote. "He can weight his own tweets and those of the sources he likes and end up with a consumer facing AI that can be a virtual Elon."
"Pretty cool," Cuban added. "Pretty scary."
In a separate note, Cuban explained what, in his view, is the biggest risk posed by this powerful technology.
"It makes dangerous people smart," he said. "And yes, there will be unintended consequences. I'm not sure what they are yet."
