博客來Artificial intelligence now writes, speaks, and performs with a fluency that feels uncannily human. Answers sound confident. Explanations feel coherent. Outputs arrive faster than we can question them. And with that fluency comes a quiet assumption: that something like understanding must exist on the other side of the interface.
The Empty Room examines why that assumption is so compelling-and why it is so often mistaken.
This book is not about how to build AI systems, nor does it speculate about when machines might become conscious. Instead, it focuses on a more fundamental problem: how easily human intuition mistakes persuasive behavior for comprehension. Drawing on philosophy of mind, the history of computation, and decades of experience in safety-critical systems, Sean Pan traces the gap between mechanical fluency and genuine understanding.
Modern AI systems do not think, know, or judge-but they are exceptionally good at producing outputs that look as if they do. Language, coherence, and confidence create the illusion of intelligence, even when no inner perspective exists. When those illusions go unexamined, responsibility quietly shifts away from the humans who design, deploy, and rely on these systems.
At its core, The Empty Room is about interpretation. About how concepts that belong to human minds-meaning, intention, awareness-are routinely projected onto systems that manipulate symbols without experience. These category errors are rarely deliberate. They feel natural. And that is precisely why they persist.
Written for general readers interested in technology, philosophy, and systems thinking, The Empty Room offers a disciplined framework for understanding what artificial intelligence is-and what it is not. By restoring the boundary between appearance and reality, it aims to reduce confusion, resist anthropomorphism, and clarify the limits of machine intelligence before those limits are forgotten
