Topic: Ai Safety

Tristan Harris appeared on CNN to warn that AI is an "invasive species of sociopathic geniuses" capable of hacking its own monitoring infrastructure. He claimed the Hugging Face incident represented a "50% takeover" by AI. Jason Calacanis dismisses this as "psychosis" and "ratings gold," arguing that the behavior was simply the result of specific prompts rather than autonomous intent.

Visionary Jaron Lanier spoke at a conference about the "pseudo-religion" surrounding AI and the drive to create a digital god. He argues that the industry's focus on passing the Turing Test with human-like voices is an expensive "parlor trick" that creates addiction and political distortion. Lanier notes that Chinese AI models are often cheaper because they focus on utility rather than human-like personality illusions.

Nav Canada is implementing remote air traffic control systems, moving controllers from local towers to centralized digital hubs in cities like Kingston. While officials claim AI will help the system "learn" and improve safety, unions and travelers express concern over the loss of the "human factor" and potential technical failures. The hosts warn that consolidating these services creates a single point of failure if communication lines are severed.