Musk’s concerns over Google DeepMind ‘AI Dictatorship’ revealed in emails from 2016 — communications released during the recent OpenAI court case

Elon Musk’s lawsuit against OpenAI has opened a can of worms, as emails between OpenAI co-founders Greg Brockman, Ilya Sutskever, Sam Altman, and Musk himself have revealed turbulent behind-the-scenes drama. According to the Transformer, an email from Musk in 2016 said, “Deepmind [sic] is causing me extreme mental stress. If they win, it will be really bad news with their one mind to rule the world philosophy.” He further aired concerns that “Demis [Hassabis] could create an AGI dictatorship.”

Demis Hassabis is one of the founders of DeepMind, an artificial intelligence research firm founded in 2010, and acquired by Alphabet, Google’s parent company, in 2014. The success of DeepMind is probably one of the major reasons behind the founding of OpenAI.

Altman emailed Musk in May 2015, saying, “Been thinking a lot about whether it’s possible to stop humanity from developing AI. I think the answer is almost definitely not. If it’s going to happen anyway, it seems like it would be good for someone other than Google to do it first. Any thoughts on whether it would be good for YC [Y Combinator] to start a Manhattan Project for AI?” Musk responded with, “Probably worth a conversation.” And, after around six months, the company behind ChatGPT was founded.

Aside from this inter-company rivalry, there’s also some tension going on behind the scenes at OpenAI. Two co-founders — Greg Brockman and Ilya Sutskever — had concerns about Altman and Musk. In September 2017, they emailed Altman asking him why he wanted to be CEO of the company and mentioned that they didn’t fully trust his judgment. Simultaneously, they were skeptical of Musk’s intentions. They said that even though he (Musk) claims to not want control of the final AGI, the negotiations revealed that absolute control was crucial to him. So, they (Brockman and Sutskever) fear that the OpenAI’s structure could allow Elon Musk to be a dictator in the company, should he decide to become one.

The Transformer says that both of these emails did not sit well with their recipients. Shivon Zillis — another OpenAI board member and a close associate of Musk — updated him regarding her conversation with Altman. She said that Sam lost trust with his two other co-founders, Greg and Ilya, through the negotiation process, adding that he felt they were inconsistent and childish. As for Elon, he replied directly to the two and said that it was the final straw for him and that the group should look for funding on its own or let OpenAI remain as a nonprofit entity.

These emails reportedly happened between 2015 and 2018, when Musk was involved in OpenAI. Although it was founded as a nonprofit organization, OpenAI’s leadership team eventually realized that it would need billions to fund the massive computer power required to train its models. They realized that they needed to convert OpenAI to a for-profit entity to secure this funding, but negotiations between Musk, Altman, and its other board members faltered when they couldn’t agree on terms.

Eventually, Musk left OpenAI and founded his AI startup, xAI, which now has a 100,000-GPU-strong data center with a 150-MW power supply. As for OpenAI’s funds, Microsoft eventually stepped in 2019, when it invested a billion dollars in the company. Today, OpenAI is one of the leaders in the field, with ChatGPT, and Microsoft is using its technology as the basis for Copilot. It also plans to invest billions more, as the software giant sees a future in AI. On the other hand, Intel got left behind after it declined to purchase 15% of OpenAI for $1 billion in 2017, saying that the company didn’t see AI providing a return on investment anytime soon.