In recent weeks, an artificial intelligence program named GPT-3 has wowed people. It’s arguably the most powerful and smart language generator ever produced. And it inevitably sparked a recurring discussion: While AI is good at specific tasks — will it ever be capable of true intelligence? That’s a very human-centered way to think about it. Why not turn things around for once, and let an artificial intelligence program argue whether or not humans are capable of true intelligence? This is exactly what you’re about to read. GPT-3 pulles no punches.

From Weekly Filet #312, in August 2020.