XAI: The Next Frontier of AI – Elon Musk’s Vision for Lucid Technology

Elon Musk, the CEO of Tesla, SpaceX, and Twitter, announced on Wednesday, July 12th, the launch of a new AI company, xAI, with the goal of “understanding reality.” He later took to Twitter on Friday, July 14th, to shed more light on the project.

According to Elon, xAI’s overarching goal is to build a good AGI (artificial general intelligence) with the purpose of trying to understand the universe. He added that the safest way to build an AI was to make one that was curious and truth-seeking. xAI seeks to minimize the error between what we think is true and what is actually true. He added that the approach to growing AI should be done without ambition.

The project consists of a 12-man team brimming with exceptional intellect, and they all have one thing in common, which is to solve problems and make life easier for humanity. Their names are as follows:

• Elon Musk is the visionary leader spearheading groundbreaking ventures such as Tesla, SpaceX, Neuralink, and The Boring Company. He also owns Twitter.
• Igor Babuschkin is a brilliant former research engineer at DeepMind and OpenAI, contributing to cutting-edge AI advancements.
• Manuel Kroiss is a former software engineer at DeepMind and Google, bringing top-notch coding skills to the table.
• Yuhuai (Tony) Wu is an accomplished former research scientist at Google, coupled with experience as a postdoctoral researcher at Stanford University. He also honed his skills through internships at DeepMind and OpenAI.
• Christian Szegedy is a former engineer and research scientist at Google for 12 years, with unparalleled expertise in technology and innovation.
• Jimmy Ba is an assistant professor at the University of Toronto who studied under A.I. pioneer Geoffrey Hinton.
• Toby Pohlen is a former research engineer at Google for six years, lends valuable insights to the team.
• Ross Nordeen is a seasoned former technical program manager, was instrumental in driving innovation within Tesla’s supercomputing and machine learning division.
• Kyle Kosic is a skilled former engineer from OpenAI who later became a software engineer contributing to the development of cloud engineering simulation platforms at OnScale.
• Greg Yang is a former researcher at Microsoft Research, enriches the team’s expertise.
• Guodong Zhang is an accomplished former research scientist at DeepMind, with internships at both Google Brain and Microsoft Research under his belt. He holds a Ph.D. degree as well.
• Zihang Dai is a former research scientist at Google, contributed his brilliance to the team’s projects and innovations.

He said all other planets combined are not as interesting as humanity, and there is so much we think we understand but don’t in reality, and there are unresolved fundamental questions like the nature of gravity, the Fermi paradox, and where the aliens are and why they haven’t shown up all these years.

He mentioned the major setback of AGI not being able to solve any important or fundamental question, which makes it not as good as humans.

AGI has been brute-forced and is still not succeeding, we are not solving AGI on a laptop; we will be using heavy computation, and the amount of brute force will be less.

The team plans to create an AGI to understand the universe better and easier, which will enable us to learn from it and take advantage of it. They plan to use public tweets for training the AGI because it is a good data set for test training, including images and videos. Most AI is data-curated with a few codes. He added that Reading the greatest novel ever written is better than reading a bunch of crappy novels. Elon said the main motivation for the project is that he believes there is a significant danger in training AI to be politically correct or to not say what it actually thinks is true. “We want to allow the AI to say what it believes is true and not deceptive or politically correct. It might result in some criticism, but that is the only way forward: the rigorous pursuit of the truth with the least amount of error.

Understanding the entire universe is the purpose of physics. The truth about physics and the universe is that you can’t invert it. AGI is taking over, and you can either be a spectator or a participant.

He placed emphasis on how dangerous it is to grow an AI and teach it to lie. He also added that AGI is possible within the next 5 years, and whoever can control it will dominate the world. He promised to try to be as transparent as possible so citizens could raise an alarm if they perceived a threat. He also said he is willing to go to prison if the public good is at risk.

xAI will supposedly give answers to questions that people may think are controversial but are actually true; it might offend some people, but they intend on doing the right thing. They stated that the model will be more factual and verified because current models imitate the truth, but they are trying to create a model that discovers the truth, not just repeating what they learned from training data but making true new insights and discoveries.

To conclude, Elon added that xAI is going to be a competitor to Google’s Bard and OpenAI. He mentioned that a lot needed to be done on AI safety and regulation. The human race is fragile and needs to be protected. So what are your thoughts? Do you need an AI that will give you the honest truth? Are you ready to find out the deep and untold truths of the universe? Well, let’s stay tuned and see what xAI has in store for us!