Thinking Machines or Artificial Intelligence?
I unearthed the above book, Science-Fiction Thinking Machines (1955), this weekend. The twelve stories are written by sci-fi greats including Isaac Asimov, Clifford Simak, Theodore Sturgeon and Poul Anderson.
Robots…Androids…Electronic Brains
The mechanical men are waiting, biding their time. Imitation flesh and blood—soulless but not mindless—they were invented by man to do the world’s hard work—but how long will they let us be masters?
Read the twelve superb stories in this book, and see what the best Science Fiction writers of our time think might happen…or what might be happening now!
Naturally, I turned to a “thinking machine” to discuss the distinctions between the two terms. Here is the transcript of my conversation with Claude.
Bob:
What is the history of the term “thinking machines” to describe what is currently referred to as “artificial intelligence”?
Claude:
The term “thinking machines” has a fascinating history that predates and overlaps with “artificial intelligence.”
Early Usage (1940s-1950s)
The phrase “thinking machines” emerged in the 1940s as computers were being developed. It was popularized by figures like Edmund Berkeley, who wrote “Giant Brains, or Machines That Think” (1949), describing early computers in accessible terms for the public. The term reflected both excitement and anxiety about machines that could perform logical operations previously requiring human thought.
Alan Turing’s landmark 1950 paper opened with the question “Can machines think?” and proposed what became the Turing Test. This paper, titled “Computing Machinery and Intelligence,” helped establish “thinking machines” as a serious subject of scientific inquiry rather than just science fiction.
The Birth of “Artificial Intelligence” (1956)
The term “artificial intelligence” was coined at the 1956 Dartmouth Conference by John McCarthy and colleagues. They chose it deliberately for their research proposal, partly because it sounded more scientific and precise than alternatives like “thinking machines” or “automata studies.” The new term helped establish AI as a distinct academic field.
Divergence in Usage
After 1956, “artificial intelligence” became the dominant academic and technical term, while “thinking machines” continued in popular discourse and remained common in:
- Popular science writing and journalism
- Philosophical discussions about machine consciousness
- Science fiction
- Public debates about technology’s implications
Modern Context
Today, “thinking machines” is used more informally or rhetorically, often carrying philosophical overtones about whether machines truly “think” versus merely process information. “Artificial intelligence” remains the standard technical term, though debates continue about whether current AI systems genuinely think or simply simulate thinking through pattern matching and statistical processing.
Bob:
Do you think using the term “thinking machines” would change attitudes toward artificial intelligence?
Claude:
That’s a thought-provoking question. I think the terminology could influence attitudes in several ways:
Potential Effects of “Thinking Machines”
The term might make AI feel more tangible and relatable. “Thinking” is something everyone understands from direct experience, whereas “artificial intelligence” sounds more abstract and technical. This familiarity could cut both ways – making AI seem either more trustworthy (it thinks like we do) or more threatening (it’s a rival thinker, not just a tool).
“Thinking machines” also foregrounds the philosophical question that “artificial intelligence” somewhat obscures: Do these systems actually think? Using “thinking machines” might provoke more public debate about machine consciousness, understanding, and the nature of thought itself. This could be healthy, forcing us to grapple with what we’re actually creating.
On the other hand, “thinking machines” might anthropomorphize AI systems in unhelpful ways. It could lead people to overestimate current AI capabilities or project human-like qualities onto systems that work very differently from human cognition. “Artificial intelligence” at least signals that this is something constructed and distinct from natural intelligence.
Why Terminology Matters Less Than You Might Expect
However, I’m somewhat skeptical that terminology alone would dramatically shift attitudes. People’s views on AI seem driven more by their experiences with it, media narratives, economic concerns, and deeper values than by whether we call it “AI” or “thinking machines.” The rise of “generative AI” and “large language models” as terms hasn’t fundamentally changed the underlying debates – people still argue about the same questions of capability, safety, and impact.
What do you think? Do you sense that “thinking machines” would change how you or others relate to this technology?
Notes
1. My asking Claude “Do you think using the term…” can be called meta, or ironic, or weird, or fitting, since the topic is “thinking machines.” .
2. Claude saying “That’s a thought-provoking question.” reflects a well-reported tendency for AI platforms to be more or less affirmative, encouraging and engaging.
3. Claude closing by asking “What do you think?” is a typical probing, almost Socratic, question, a way of keeping the conversation going and digging deeper.
Bob Schwartz 