OpenAI researchers warn of a “model threatening humanity”

Estimated read time 2 min read

  OpenAI researchers warn of a

The reasons that motivated the dismissal of the CEO of OpenAI last week are beginning to emerge. And all this looks like a real science fiction scenario. A powerful AI model, threatening humanity, and created by several OpenAI researchers, would be the cause of the case.


According to The Information, a group of OpenAI researchers sent a letter to the board of directors a few days ago, mentioning the existence of a generalized artificial intelligence (AGI) with superpowers. The letter was sent shortly before the meeting that resulted in the dismissal of Sam Altman.


According to the article, the AI model is called Q*, pronounced “Q Star”. This AI would evolve tens of times faster than existing models. The researchers claim that they were shocked by its performance after testing the mathematical abilities of the model.

Q* is very different


Generative AI, such as ChatGPT, generates content by predicting the next word based on statistics and probability. And it can only provide results in the context of data learning. Concretely, this means that the AI cannot answer questions about data that it does not know.

But it seems that Q* is very different. This AI can understand and solve computer problems without learning additional data. That is why he is able to solve new computer problems that do not appear in the set of data that he has swallowed. Concretely, like a human, this AI can apply its existing knowledge to solve new problems.


To the layman, this ability may seem simple. But for AI researchers, this is a technological breakthrough. “Q-Star’s ability to solve mathematical problems on its own is a sign that we are getting closer to AGI,” explains The Verge.

Vigilance about the pace of development


In their letter, the researchers warned the board of directors that “the Q-Star model could be used as a system powerful enough to destroy humanity” and that it should be “vigilant about the pace of development”.

“OpenAI can lift the veil of the unknown and push the limits of the
technological progress” had declared the same week Sam Altman at the summit of the
Asia-Pacific Economic Cooperation (APEC), shortly before being fired, adding that he was “honored to lead the company at this
moment of its history”.

To go further on the news of OpenAI


Source: “ZDNet Korea”

You May Also Like

More From Author