It's intriguing to see Andreessen express a desire to create 12 Elon Musks, given that Elon is one of the most vocal advocates for AI safety. Elon has consistently expressed concerns about the unchecked development of AI, and he has called for a pause on large-scale AI systems. It would be interesting to see Andreessen address his disagreement with Elon directly, rather than through hand-wavy dismissal of AI existential risk. Of course, sometimes Elon says crazy things, but based on this interview, it certainly looks like Elon has thought more seriously about this issue than Andreessen.

Expand full comment

I went through the transcript looking for any mention of the possibility that AI that's smarter than human, might end up running the world on its own, instead of being in a "symbiosis" with humans. I didn't see any mention. So, we still don't know what Marc thinks about this. Would he take a Yann LeCun line, that AI will remain subordinate to humans indefinitely, despite being more capable? Or does Marc regard AI takeover as one of the genuine risks, but a risk worth taking?

Elon just launched his own ChatGPT rival (Grok), so it's unclear what he actually thinks about a pause.

Expand full comment

I'm aware of no one who has be more clearly concerned with existential AI risk than Elon. Here is the letter Elon signed in 2023 advocating a pause.


Here is Elon in 2014 declaring AI the most serious threat to the survival of the human race.https://www.theguardian.com/technology/2014/oct/27/elon-musk-artificial-intelligence-ai-biggest-existential-threat

Here is Elon in 2017 "calling regulation generally "not fun" and "irksome," but he said that in the case of AI, the risks are too high to allow AI to develop unfettered."


The current capabilities of AI Grok may not be advanced enough to breach the proposed pause. In any case, Elon is not necessarily bound by a ban he supports, especially if it lacks widespread agreement. Cynics may argue that Elon's advocacy is a strategic move to delay competitors while he advances his own AI technology. However, considering Elon's long standing concern for the existential risks of AI, I think his advocacy is genuine.

Andreesen, on the other hand, repeatedly displays dismissive contempt for people concerned with AI existential risk. I would love to be convinced that Andreesen is right and Musk is wrong, but I've seen no evidence that Andreesen has given existential AI risk any serious thought.


Expand full comment