Bonus: the Strange Case of Dr. Elon and Mr Musk

Is the master of distraction actually the spider in the AI web? Or is he just an opportunistic egomaniac?

This is a bonus newsletter in our series about The Deep Weirdness of Artificial Intelligence. Previous reports in this series have focused on several aspects of the rushed deployment of consumer-facing AI apps that have stirred up fear, uncertainty and doubt.

Running through the story like a bright red thread is the name of tech titan Elon Musk.

His objectives are conflicted. He seems to be working on all sides of the AI controversy.

Is he for AI or against it? Or both?

What’s Elon Musk’s AI gambit?

Business Insider reports that Elon Musk recently spent tens of millions of dollars to purchase 10,000 GPUs for a generative AI project inside Twitter. That is approximately the same number of chips as OpenAI uses to run GPT-4.

In late February, Musk conferred with former DeepMind researchers about launching a new AI lab to compete with OpenAI’s ChatGPT.

It is no longer a secret that he is building a new company in Nevada called X.ai.

Hang on. There is something weird about Elon Musk’s leap into the AI arms race.

Musk claims that he fears AI might harm humans… and yet, at the same time, he also makes boastful claims about the AI already deployed in Tesla products that seem to have been released before they were entirely safe.

Musk funds research and development of large language models… but he also donates to the think tank that advocates against the AI rollout.

Let’s break this down.

Q: Isn’t Elon Musk an outspoken critic of artificial intelligence?

A: Yes, he is.  For years he has used Twitter as an outlet for AI doom mongering. He said that developing AI is like “summoning the demon” and claimed it was “worse than nuclear weapons”. 

In 2017, Musk told U.S. state governors that “AI is the biggest risk that we face as a civilization” and called for regulation.

Q: But wait, doesn’t Tesla use AI in cars?

A: Yes. The auto manufacturer maintains an active AI development program and has aggressively recruited top tier AI researchers.

During the past two years, Tesla has celebrated an “AI Day” to showcase the company’s progress on artificial intelligence. Tesla vehicles equipped with Autopilot currently use an onboard AI system to process data gathered by eight vehicle cameras to generate a real-time 3D model that identifies other vehicles, pedestrians, lanes, roads, traffic signals and obstacles.

Tesla also claims to be building an autonomous humanoid robot.

Q: Has Tesla autopilot been implicated in any fatal accidents?

A: Yes. At least 14 motor vehicle fatalities involving Tesla autopilot have been reported, more than all other electric vehicles combined.

Musk’s doomsaying about AI seems to be a kind of psychological projection. The fact remains that Tesla AI products are implicated in more deaths than any those of other AI organization.

In terms of artificial intelligence that causes actual harm to human beings, it can be said that Tesla is the reigning world champion of killer AI.

Q: Have Musk and/or Tesla made misleading statements about the AI capabilities of Tesla self-driving cars?

A: According to shareholder lawsuits, customer lawsuits, the California Department of Motor Vehicles, and the German government the answer seems to be consistently yes.

Q: Are Tesla and Musk under investigation by the US federal government for claims made about the self-driving capabilities of Tesla cars?

A: Yes. At the request of the National Highway Traffic Safety Administration, the US Department of Justice has requested documents from the Securities and Exchange Commission in connection with an investigation into claims made about the self-driving Tesla vehicles.

According to the New York Times:

The National Highway Traffic Safety Administration, the primary auto safety regulator in the United States, is investigating Autopilot after becoming aware of 35 crashes involving the system, including nine that resulted in 14 deaths. Its investigation covers 830,000 vehicles sold in the United States and is looking at Full Self-Driving as well as Autopilot

Tesla confirmed that it ““has received requests from the DOJ for documents related to Tesla’s Autopilot and FSD features.”

Q: Was Musk also an early investor and co-founder of OpenAI? 

A: Yes, along with fellow PayPal founders Peter Thiel and Reid Hoffman, Musk invested in OpenAI

Q: Did Musk attempt to take over OpenAI?

A: Yes, according to Semaphore, in 2018, after a bungled a takeover attempt, Musk was pushed out of OpenAI by the other co-founders. They were already frustrated with his attempts to poach OpenAI researchers for Tesla’s autopilot project.

Musk had previously pledged $1 billion in funding for OpenAI. But after he parted ways in a huff, he reneged on that commitment. That’s what caused CEO Sam Altman to seek strategic funding from Microsoft.

This sequence of events led OpenAI to becoming a for-profit company and a provider of AI technology to Microsoft.

Naturally, Musk is complaining about that, too.

Q: Did Musk also contribute funding to the Future of Life Institute?

A: Yes. Musk is a longtime fan and supporter of FLI founder Nick Bostrom, whom he has quoted on Twitter.

Q: Did Musk sign the open letter from The Future of Life Institute that called for a six-month pause in AI research?

A: Yes

Q: Is is possible that Musk’s call for a temporary pause in AI research was just a cynical ploy to buy time for for Twitter’s AI project to catch up with OpenAI?

A: Seems likely. Not publicly confirmed.

This is what Musk’s friend and former colleague Reid Hoffman indicated when he mused about Musk’s motive for signing the letter: "I think some of it's a little bit less well intentioned, like 'Everyone else slow down, so I can speed up.''

Q: So what’s Elon Musk really like as a person?

A: According to OpenAI co-founder Sam Altman, Musk is a jerk. “He has a style that is not a style that I’d want to have for myself,” Altman said on Kara Swisher’s podcast.

There is a pattern in his behavior. When Musk tweets his concerns about AI, what he really seems to be saying is that he has no confidence in anyone else’s ability to manage it responsibly. And then he carves out an exception for himself.

Musk fundamentally believes that no one else can handle complex technology as well as he can. And he seems determined to soak up all of the oxygen in the room before anybody can argue otherwise.

---