Corporate betrayal in the AI industry

AI engineer Xuechen Li became the center of the biggest corporate scandal in the AI industry. Elon Musk’s company sued its former employee. And insists that the xAI engineer stole Grok secrets for OpenAI. The story developed quite interestingly. In early summer, Li received an offer from OpenAI. Accepted the offer and immediately sold $7 million worth of xAI stock. Classic corporate betrayal scheme. But remained working at Elon Musk’s company. Already sounds strange.

Then in July the key moment occurred. Li accidentally gained access to confidential xAI files. According to the company’s claims, this was information about advanced AI technologies surpassing ChatGPT. And there were technological secrets capable of changing the balance of power in the market. Accidentally, sure.

On August 14, an internal meeting took place. Where Li confessed to stealing some files. But tried to cover tracks of his activities. Investigation revealed additional NDA materials on his devices. And he kept silent about them during his confession.

xAI now demands from the court compensation and prohibition on Li’s transition to OpenAI. Because stolen materials could allow OpenAI to improve ChatGPT. Implement more creative and innovative features from xAI. And this would be a direct threat to competitive advantage.

Li worked as an engineer at xAI and personally participated in Grok’s training and development. Had access to critically important information. Knew the internal system architecture.

The engineer “accidentally” gets access to secret files. “Accidentally” copies them and “accidentally” forgets to mention it. After previously “accidentally” selling $7 million worth of stock. And accidentally forgets to resign for two months. Amazing number of “accidents”!