Before the term “AGI” turned into a Silicon Valley war cry, before it was the centerpiece of OpenAI board meetings and trillion-dollar valuations, there was a man in a dim university basement, typing out a paper that no one thought would matter. His name was Mark Gubrud, and the idea he put into words — artificial general intelligence — would end up reshaping the entire conversation around what it means to build a thinking machine.
The Basement Where the Future Was Named
It was 1997. Gubrud, a graduate student at the University of Maryland, sat in the sub-basement of the chemistry building, half-listening to a noisy sump pump kick on and off. In that strange, echoing space, he began writing about nanotechnology and military risks — a paper that would become “Nanotechnology and International Security.”
Buried inside it was a phrase no one had seen before: “advanced artificial general intelligence.”
Gubrud used it not to excite people, but to warn them. He described AI systems “that rival or surpass the human brain in complexity and speed… that can acquire, manipulate, and reason with general knowledge.” To him, this wasn’t a cool future — it was a potential arms race.
The irony is that this quiet academic paper, written in warning, ended up defining the goalpost for the entire AI revolution.
How a Cautionary Term Became a Rallying Cry
The word “AGI” disappeared for a few years, buried in conference proceedings. But as the AI field woke up in the 2000s, researchers like Ben Goertzel and Shane Legg began searching for a way to describe human-like machine reasoning. “Strong AI” sounded clunky. “Real AI” was vague.
Then the old phrase resurfaced. “Artificial general intelligence” had the right rhythm — specific enough for science, yet broad enough to dream on. And so it stuck.
Soon, it wasn’t just a term. It became a mission statement. DeepMind wanted to “solve intelligence.” OpenAI promised to “achieve AGI safely.” The world began to race toward a concept that was never meant to be a finish line.
The Forgotten Author
When Wired recently tracked Gubrud down, he spoke with the quiet exhaustion of someone who’d seen his words run away from him. “I’m 66 years old,” he said. “No name, no money, no job. But I guess I invented AGI.”
That sentence sums up the paradox of modern tech history: the people who name the future rarely profit from it.
Gubrud’s original paper wasn’t about building AI at all. It was about controlling it — about the danger of combining self-learning machines with military power. Yet the term he coined now drives the very competition he hoped to restrain.
The Shift in Meaning
When Gubrud wrote about AGI, it was a philosophical and ethical concept. He warned that nations could use intelligent machines to automate not just industry, but war itself. It was a call for global coordination — not a race.
Two decades later, AGI is shorthand for the ultimate tech trophy. It’s the “moonshot” every lab wants to claim. Investors see it as destiny; governments see it as power.
And so, the same idea that once stood as a warning has become a corporate ambition.
Why His Voice Still Matters
The world Gubrud feared is almost here. AI models can already generate code, simulate reasoning, and optimize themselves. The line between “tool” and “mind” is blurring.
His early papers feel prophetic now. He described the possibility of autonomous systems that evolve beyond human oversight, and questioned whether nations could handle the speed of such change. That conversation — safety, ethics, control — is finally mainstream.
Yet his name still isn’t.
What Journalism Can Learn from Gubrud’s Story
Tech reporting in 2025 moves fast — sometimes too fast. We cover AGI like it’s a finish line, but forget to ask who drew it. Gubrud’s story is a reminder that the language we use carries moral weight.
Three takeaways stand out:
- Definitions drive direction. “AGI” started as a caution and became a goal. The shift happened because no one guarded its meaning.
- Forgotten thinkers still shape the present. Gubrud didn’t fade because he was wrong — he faded because hype moves louder than humility.
- Caution and curiosity must coexist. Progress doesn’t have to mean recklessness, and reflection doesn’t have to mean fear.
The Legacy in the Noise
Mark Gubrud never built an AGI system. He never filed a patent or founded a company. But his words — tucked into a paper about nanotech and national security — gave the modern AI world its most powerful term.
In a sense, he didn’t just invent a phrase. He named a future. And that future, with all its beauty and chaos, is now rushing to meet us.
Maybe that’s the quiet poetry of his story: that the man who tried to warn humanity about intelligent machines accidentally gave those machines their mission statement.
And somewhere in a quiet Maryland basement, the echo of that idea still hums — steady as a pump switching on, shaping the rhythm of everything that came after.
Visit: AIInsightsNews
