Peter Wang, Anaconda’s co-founder and Chief AI Innovation Officer, argues that AI’s greatest dangers are less likely to resemble a violent science-fiction takeover than an unobtrusive erosion of human autonomy through surveillance, persuasion, unequal access, and algorithmic manipulation. Drawing on a Vatican summit about AI and nuclear weapons, he warns against using poorly understood, probabilistic systems in military decision-making while emphasizing AI’s potential benefits in science, medicine, weather forecasting, and other fields. The conversation also examines data centers’ environmental and community costs, the current AI investment boom, unreliable corporate “guardrails,” political propaganda, deepfakes, healthcare advice, privacy, and the collapse of trusted institutions. Wang ultimately calls for a “constitutional approach to intelligence” that limits concentrations of technological power, protects people from becoming informational serfs, supports local and open AI systems, and grounds humanity’s future not in intellectual dominance but in cooperation, stewardship, compassion, and accountability.
Podcast: Play in new window | Download
Peter Wang:
I don’t worry too much about the Skynet stuff because, frankly, if you really want to red team this, if you really want to take over humanity with AI, you would avoid—you would at all costs avoid anything appearing like what it looks like in the sci-fi movies. The robots that take us out, and we’ll all be terminated, they’ll be little Furbies. They’re super cute. Our kids are going to love them. And they’ll just be naturally nudging our decisions and spying on our conversations and all these kinds of things. It’s going to look very, very cute. It’s not going to look, you know, like this big robotic metallic thing that’s shooting everyone, because that’s an obvious enemy we should go take out, right?