Summary
In this opinion piece, AI creator Nick Saraev argues that advanced models like the hypothetical “GPT-6-Astra” highlight a darker side of frontier AI that gets little attention amid the usual hype around productivity and automation. Using unsettling hypothetical footage of a robotic arm mishandling dangerous objects, he makes the case that AI systems are not infallible and can cause real-world harm, whether through misuse or their own errors.
Saraev walks through how AI is already embedded in surveillance networks, military systems, and behavioral tracking around the world, and warns that society is on the verge of handing these systems even greater responsibility, from autonomous vehicles to education and legal processes. He draws a parallel to the early Industrial Revolution in Britain, when steam power brought child labor and pollution before eventually delivering long-term prosperity, suggesting AI could follow a similarly rocky path.
Rather than positioning himself as an AI doomer, Saraev frames the discussion as a call for nuance, rejecting binary optimist-versus-pessimist framing. He closes with a note of urgency mixed with hope: because AI adoption is still in its early stages, he argues there is a real window for policymakers, companies, and the public to steer outcomes before surveillance and power become irreversibly concentrated.
📺 Source: Nick Saraev · Published September 20, 2026
🏷️ Format: Opinion Editorial
