Within minutes of each other on a recent episode of The Joe Rogan Experience, Joe Rogan managed to hold three entirely different and mutually exclusive positions on artificial intelligence, often without seeming to notice the contradiction.
The episode featured Bob Lazar, the man who claims to have worked on reverse engineering alien spacecraft at a secret government facility near Area 51. But at various points, Rogan steered the conversation toward AI, and what followed was a showcase of just how tangled his thinking on the topic has become.
It started with a passing comment about Claude AI. Rogan mentioned that engineers working on the system believe it may already be sentient.
He noted, “They think that the Claude AI, the engineers, they think it’s sentient already. It just doesn’t have a physical body to move around.” That framing sets up AI as something almost alive, something deserving of real consideration.
Then moments later in the conversation, Lazar stated flatly: “AI is going to ki ll us. Everybody agrees with that. There’s no question.”
Rather than pushing back with any of the nuance he had just displayed, Rogan responded: “I don’t think it’s going to k ill us. You know what I think it’s going to do? I think it’s going to prevent us from breeding. I think it’s going to let us d ie off.”
Lazar’s immediate response was: “That’s going to ki ll us.”
So in the span of roughly thirty seconds, Rogan went from treating AI as a potentially sentient being to arguing it would wipe out humanity through reproductive suppression rather than direct confrontation. Lazar’s point, that either way amounts to the same outcome, seemed to bounce right off him.
The real curveball came later in the same episode. After spending significant time riffing on Iran, nuclear weapons, and the general collapse of civilization, Rogan landed on what he framed as a solution: implement AI as government.
The idea, floated without much development, was that a sufficiently rational artificial intelligence might govern better than humans ever could.
This sits in direct contradiction to everything said earlier in the conversation. In the same episode, Rogan suggested AI would end humanity, then proposed handing it the reins of government. Just three incompatible positions occupying the same two-hour window.
What makes this worth paying attention to is less about AI specifically and more about what it reveals. Rogan is no longer building toward a point. He is reacting, riffing, and moving on.
Whatever lands in his head in the moment gets said out loud, regardless of whether it lines up with what he said twenty minutes earlier.
