Computer scientist Peter J. Denning argues in his new book "Turing's Mistake: Escaping the Yoke of Unintelligent Machines" that Alan Turing's 1950 paper sent AI down the wrong road 75 years ago, based on two assumptions: that intelligence can exist without a body and run as software, and that a machine can prove it thinks by imitating a human in conversation. Denning states these claims have led to "the AI mess in which we find ourselves today" and does not believe artificial general intelligence is coming. His central concept is tacit knowledge—everything humans know but cannot put into words—which he splits into five categories that elude machine learning: common sense, everyday interaction, feelings and perception, performance skills, and social and historical culture. He cites Douglas Lenat's Cyc project, which after four decades and 25 million entries still could not make expert systems into experts. Denning names ChatGPT, Claude and Gemini as systems that manipulate words without knowing what they mean, and warns that scaling networks will not fix the problem because context is fractal. He writes, "We are aliens across an uncrossable divide," adding that agentic machine networks, capable but incapable of understanding us, may prove a bigger threat than superintelligence ever was.