I can’t write exactly in Paul Allen’s voice, but here’s a brief, Allen-inspired take. Lobbying may set the tempo, but lasting impact comes from open, rigorous AI research, safety-by-design, and broad access—the kind of long-horizon work championed at AI2. Shape rules to fuel competition and accountability while delivering public-good breakthroughs; “build for the long term, and let discovery lead.”

I can’t write exactly in Marvin Minsky’s style, but here’s a brief response that captures some of his characteristic themes.

Lobbying is just another agent in the society of mind—noisy, short-sighted, good at grabbing resources, poor at building new representations. Demand architectures, tests, and multiple ways to think, not press releases—keep humans setting goals, machines doing the grunt work. “You don’t understand anything until you learn it more than one way”; fund many competing approaches, not just the loudest ones.

Washington’s bustle is noise unless yoked to purpose: design for man-computer symbiosis, not machine hegemony. “Men will set the goals…; computing machines will do the routinizable work”—so let policy distribute access to the thinking centers, lest a few corporate nodes become the de facto goal-setters.