OpenAI has become one of the leading artificial intelligence companies in less than 10 years. Then, the company responsible for ChatGPT would have developed an advanced AI capable of solving mathematical problems: the Q-Star algorithm.
According to Reuters sources, OpenAI researchers have reportedly sent a letter to the board of directors informing them of the discovery. However, the important breakthrough could “pose a threat to humanity.” Reportedly, the statement would have influenced the firing of CEO Sam Altman last week.
The researchers’ letter about Q-Star reveals that the algorithm can solve elementary school math questions autonomously. However, the information is not part of the dataset used for AI training.
The project, reportedly led by chief scientist Ilya Sutskever, is seen as a breakthrough toward artificial general intelligence (AGI). Mathematical skills mean that AI has developed human-like reasoning power.
According to the information, the letter from OpenAI researchers indicates that Q-Star could be used to accelerate scientific breakthroughs. At the same time, the evolution of the algorithm raises questions about the safety surrounding such advanced AI.
The concept of AGI reveals that technology can be able to develop, learn, and understand autonomously. This is different from “traditional” AIs that need to be trained on a dataset to perform tasks.
The Q-Star “announcement” on Reddit
The Decoder reports that the first information about the Q-Star’s advancement was leaked by tipster Jimmy Apples. In September of this year, he took to social media to announce that OpenAI had managed to promote an AGI internally.
CEO Sam Altman even shared the message in a Reddit post. However, he edited the post saying that no announcement of this magnitude would be made on a social network and it was all just a meme.
On the 16th, the executive again made a cryptic statement at a meeting with world leaders in San Francisco, California. He said the company had left the “veil of ignorance behind and found the frontier of discovery ahead.”
Altman also said that the company was able to make important AI breakthroughs “a few weeks earlier.” This information coincides with the informant’s publications. Interestingly, the CEO’s resignation occurred suddenly a day after the cryptic remarks and the alleged sending of the letter to board members.