Elon Musk remains with his guard up in front of Sam Altman. Former partners in OpenAI, today they are rivals, with the South African concerned about the rise of Artificial Intelligence, beyond ChatGPT.
A new opinion from the head of X, SpaceX, Tesla and other companies was recently known about what Sam Altman is doing with AI.
Musk supports a theory that says that Altman is “stealing the Internet” to sell it to users with ChatGPT. He did so in a tweet from a user who considers this true, the tycoon stating “that there is some truth.”
According to critics of Sam Altman, including Elon Musk, the head of OpenAI extracts and compresses data from the open (and free) Internet to build powerful Artificial Intelligence models (with which he makes a lot of money).
Altman’s strategy, always following the guidance of his critics, is to sell this data to users through APIs and impose legal barriers to prevent others from doing it too.
An API is an application programming interface that allows products and services to communicate with others, without needing to know how they are implemented. They are considered contracts of agreements between parties.
In April of this year it was learned that OpenAI had access to the old Twitter, today X, to use the data of its users in order to train Artificial Intelligence. For them he paid 2 million dollars to Elon Musk, but the magnate considered it insufficient, preventing him from continuing to do so.
Countries like Italy and China have placed restrictions on ChatGPT and similar artificial intelligences.
Seeing a lot of confusion about this, so for clarity:
openai never trains on anything ever submitted to the api or uses that data to improve our models in any way.
—Sam Altman (@sama) August 15, 2023
Sam Altman, last August, defended his company by saying that does not use data sent through its API to train its Artificial Intelligence models.
He did acknowledge that they only incorporate user-specified data into future training runs, but always with the consent of customers and to improve the model in specific ways.