Threatening AI discovery: OpenAI researchers warned board

Before Sam Altman was fired, OpenAI's research team apparently warned the board of a threatening AI discovery. According to an internal letter, this represents a threat to humanity. The discovery led, among other things, to Altman's dismissal.

Before OpenAI fired its co-founder and CEO Sam Altman for four days, the company's research team apparently wrote a letter to the board. In it they warned of a powerful AI discovery that poses a threat to humanity.

This comes from a report by the news agency Reuters citing two insiders. The letter, which has not yet been made publicly available, led to Altman's dismissal. After employees and investors took to the barricades, Sam Altman is now back in the executive chair.

OpenAI: AI discovery poses threat to humanity, researchers say

According to the insiders, the letter was part of a list of complaints from the board. Among them: concerns about commercializing AI before the consequences are clear. Loud Reuters OpenAI declined to comment.

However, in an internal message to its employees, the source said the company spoke of a project called “Q*” (pronounced Q-Star) in connection with the weekend's events. The model could therefore be a breakthrough in the area of ​​Artificial General Intelligence (AGI).

Project “Q*”: Has OpenAI developed a super AI?

OpenAI defines AGI as an AI system that can go beyond its data sets and outperform humans. The model is also able to solve mathematical problems. Researchers have so far seen mathematics as a limit to the development of artificial intelligence.

If an AI were able to solve mathematical problems, it could develop a thinking ability similar to that of humans. Such an artificial intelligence is capable of learning and understanding. In their letter, the OpenAI researchers apparently again warned the board of the dangers of AGI.

According to Reuters, Sam Altman announced at a recent conference in San Francisco that major advances in AI are on the horizon. Just one day before his two-term release he said:

Four times in OpenAI's history, most recently in the last few weeks, I have been privileged to be there as we lift the veil of ignorance and push the boundaries of discovery, and doing so is the professional honor of a lifetime.

Also interesting: