Challenge OpenAI, Musk announced the formal establishment of xAI: the goal is to "understand the true nature of the universe"!

In the early hours of July 13, Beijing time, Musk announced on Twitter: "xAI is officially established, to understand reality." Musk said that the reason for launching xAI is to "understand the true nature of the universe." It has been half a year since Ghat GPT was born, and the "Hundred Models War" at home and abroad is intensifying. What do you think of the current situation and development of AI large models?

1. Musk officially announced that xAl is hard and rigid OpenAl

The "Musk wants to toughen OpenAl " that has been rumored in recent months has finally come true. On July 12, Musk announced his latest artificial intelligence company xAl on Twitter .

The goal of the new company sounds very metaphysical: " Understanding the true nature of the universe ", but this is also very much in line with Musk's usual style of boasting.

2. 1/3 of the founding team members are Chinese

At present, the official website has announced the other 11 start-up members . They have working backgrounds in DeepMind , OpenAl , Google Research , Microsoft Research and Tesla . Outstanding achievement. It is worth mentioning that 4 of these 11 people are Chinese , accounting for 1/3 of the team . They are Yuhuai Wu , Greg Yang , Guodong Zhang and Zihang Dai . Public information shows that they have study backgrounds such as Tsinghua University , Zhejiang University , Harvard University , University of Toronto , and Carnegie Mellon University .

3. Musk played a game of chess

According to the official website , xAl is an independent company, but it will maintain close cooperation with Twitter's parent company " X Company ", Tesla and other companies. In fact, the "close cooperation" between xAl and Twitter has quietly started. Jim Fan, a senior AI scientist at Nvidia , recently revealed on Twitter that xAl is currently the only artificial intelligence company that can grab information on Twitter for training. Do you still remember the current limitation incident caused by Musk in order to curb other AI start-ups from coming to Twitter to "scrape data"? Now it seems that there is a Al baby waiting to be fed at home. In the future, high-quality big data will only be fed to my own Al language model. Moreover, Musk has used such ruthless moves to actively limit the flow of the platform, which also shows that he is ready to fight his opponents to the end. And bear the brunt of it is OpenAl . Since OpenAl launched ChatGPT at the end of last year , Musk has started to talk about the " Al threat theory ", calling out OpenAl in various ways . It seems that all this is paving the way for the launch of Al company.

Let's take a look at how Musk played this game of chess?

• February

For the first time on Twitter, he proposed to consider building "TruthGPT" to fight against ChatGPT, saying that it will be an artificial intelligence that seeks the truth to the greatest extent.

• March

Musk signed a joint letter with other industry leaders calling for a moratorium on the development of artificial intelligence more advanced than GPT-4 for at least half a year.

• April

He bought 10,000 high-performance GPU chips from Nvidia. At the same time, Musk continued to publicly criticize OpenAl for "training artificial intelligence to lie", accusing it of deviating from the purpose of "open source" and being willing to be controlled by Microsoft. money making machine. Musk made a look of being at odds with OpenAl.

In fact, he himself was one of the earliest promoters of the company, and it was not until 2018 that Musk left the OpenAl board of directors. According to media reports, on the one hand, because of differences in the development direction of OpenAl, Musk hopes to maintain the non-profit nature and continue to explore safer and more beneficial Al, while other partners hope to raise funds and realize Al commercialization as soon as possible; on the other hand, Tesla There is a conflict of interest between La La’s in-depth research on AI technology for autonomous driving and OpenAl. From this point of view, it seems that Musk can only go into battle in person to prevent Al from going on a "wrong path" to destroy mankind. Musk once said when talking about his vision for "TruthGPT": I think "an Al who cares about understanding the universe is unlikely to destroy humans, because we are an interesting part of the universe", which may be why Musk gave his Al company Set the goal of "understanding the true nature of the universe."