Key Information Points Summary from Elon’s Speech in World Artificial Intelligence Conference

Elon did not disappoint this year at the World Artificial Intelligence Conference, and even revealed some new information. Here are the key takeaways from the 5 Q&A sessions:

• Tesla is establishing an engineering team in China for Autopilot autonomous driving development. Tesla will focus on original engineering development, rather than simply copying American designs to China.

• Elon is confident that basic L5 autonomous driving functionality can be achieved this year.

  There are currently no fundamental challenges to achieving L5 autonomous driving. However, there are a lot of detail-oriented issues that need to be addressed. Tesla is currently focused on handling these details. Elon believes that these issues can be resolved based on the hardware versions currently installed in Tesla vehicles. We just need to improve the software to achieve L5 autonomous driving!

• Advanced imaging recognition systems are better than humans nowadays. The problem lies in the amount of computing power, computers required, and duration for training perception ability. How efficient is the training system for image recognition? For a given byte stream, can AI accurately recognize and process image or voice?

• The three pillars of AI and robotics technology are “perception”, “cognition”, and “behaviour.” “Cognition” is currently the weakest field.

• It is expected that Tesla’s factories will make greater use of AI and more intelligent software in the future, but it will take some time to effectively utilize AI.

This article is a translation by ChatGPT of a Chinese report from 42HOW. If you have any questions about it, please email bd@42how.com.