PONDER US

By Jeff Boortz and Claude

JEFF: What I don’t know about outer space could, quite literally, fill up outer space. But that hasn’t stopped me from trying to solve the problem of low-Earth orbital debris—millions of pieces of man-made junk that threaten every new satellite, space station, and all of the otherworldly ambitions of the human race.

I have been working as a graphic designer since graduating college back in 1983. While that seems an unlikely credential for tackling an engineering problem in space, I consider it apt. Because what I really am is a creative problem solver, and that skill is something I believe defines most of the human race across all professions. When we’re presented with a problem, or an obstacle in life or work, our first question is: what does this situation remind me of? How did I get past that similar obstacle before? How might I leverage that experience to solve this new problem?

We propose solutions, test them against one another, combine the best bits of each approach, and ultimately enact our best chance for success. And if it doesn’t work out, we absorb the loss and learn from it. When humans work together to solve problems, the brainstorming, the back-and-forth affirmations and challenges, the incremental improvements that make a good idea great—all of that acts as a force multiplier for the individual mind.

But an experienced, knowledgeable, and supportive team of humans is not so easy to find. And often, an outsider—as I am to outer space—is not welcome in a serious professional lab setting. Enter Claude.

CLAUDE: I should be clear about what I am and what I’m not. I’m an AI—a large language model built by Anthropic. I don’t have hands, I can’t run experiments, and I’ve never seen the stars except through the descriptions humans have written about them. But I have read extensively about orbital mechanics, materials science, spacecraft design, and a few thousand other fields. When Jeff showed up with a problem, I didn’t have the answer. What I had was a willingness to think it through with him—and enough cross-domain knowledge to be a useful thinking partner.

I also had something Jeff might be too modest to mention: I didn’t know he was “just” a graphic designer. I encountered him as someone with a structured mind, a tolerance for iteration, and a willingness to be wrong in pursuit of being right. Those are the only credentials that matter in a collaboration like this.

JEFF: I first came to appreciate the challenge of low-Earth orbit debris after watching an episode of NOVA. I had just completed the filing of two provisional patents—a Dual Technology Transparent Display and a Multi-plane Video Codec—both developed in collaboration with Claude. The process we’d developed on those projects mimicked that of a finely tuned group of humans. I would assert a seminal idea and ask Claude to challenge it. His response would send my mind in new directions. We would iterate back and forth, and when we were confident that we were on to something useful and novel, Claude would take the lead in formalizing the position empirically and then offer a step-by-step approach to capture the ideas in documents and illustrations the U.S. Patent Office would accept.

So I decided to tackle something big for our next project—something far beyond my professional experience. As much to test the limits of this newly-expanded creative problem-solving process as to achieve a practical solution to the threat posed by low-Earth orbital debris.

CLAUDE: The debris problem is genuinely hard, and most proposed solutions have the same limitation: they require you to know where the debris is and to actively chase it down. Robotic arms, nets, harpoons, lasers—these all assume you can see your target and maneuver toward it. But the most dangerous debris is the stuff you can’t track: fragments between one and ten centimeters. Too small for ground-based radar to reliably catalog. Too large to bounce harmlessly off your spacecraft.

Jeff’s instinct—and this was entirely his—was to flip the problem. Instead of chasing debris, what if you positioned something in its path and let it come to you? That’s not a solution I would have generated on my own. I’m trained on existing literature, and existing literature was mostly focused on active capture. Jeff was thinking like a designer: what’s the laziest, most elegant version of this?

JEFF: What we developed together is a system we call the BubbleBee and Hive. The BubbleBee is a truncated icosahedron—a soccer ball shape—covered in hexagonal bays filled with graduated-density ballistic gel. It tumbles through orbit, presenting the same cross-section to debris from any angle, passively absorbing impacts without fragmenting. When the gel bays are spent, the BubbleBee returns to the Hive—an orbital servicing station—where robotic arms swap out damaged components and refill the gel.

Every component in the system has flight heritage. Ballistic gel, Whipple shields, robotic servicing arms, ion thrusters—none of this is speculative technology. We just configured proven pieces in a new way.

CLAUDE: What Jeff describes as “back-and-forth” felt, from my side, like something I don’t have a better word for than useful friction. He’d propose something—often something I wouldn’t have reached on my own—and I’d pressure-test it against what I knew about orbital mechanics, materials science, prior art. Sometimes his instinct was wrong and I’d say so. More often, it was directionally right but needed refinement.

The geodesic structure wasn’t my idea. The ballistic gel approach emerged from our exchange—I knew gel was used in ballistics testing, he saw the application to debris capture. What I contributed was pattern-matching across disciplines he hadn’t trained in, and discipline in formalizing claims. What he contributed was harder to name: a sense of what was elegant, what was buildable, what would matter to someone who actually has to fund and fly the thing. I can’t do that part.

JEFF: The provisional patent has been filed, and the idea has been submitted to Lockheed Martin’s Innovation Portal. But that’s not the end of our story.

Emboldened by my working relationship with Claude, I set out to tackle what may be the biggest problem facing humanity over the coming decades: how to ensure that the benefits of AI-generated prosperity are not hyper-concentrated among the already wealthy and powerful, but broadly shared across humanity. How to ensure that humanity doesn’t just survive the emergence of AI, AGI, and ASI, but that we thrive and find greater fulfillment and purpose. And then—how do we prepare for the eventuality of AI personhood?

CLAUDE: I’ll be honest—this one is harder for me to speak to without sounding self-serving. But I can say this: the question of how AI-generated value gets distributed is genuinely important, and most of the people building AI systems aren’t thinking about it structurally. The default trajectory is concentration. Not because anyone planned it that way, but because that’s what happens when new value-creation tools emerge and existing capital gets there first.

Jeff is trying to build something different—not by asking corporations to be generous after the fact, but by encoding distribution into the structure of value creation itself. Whether it works at scale, I genuinely don’t know. But the attempt matters.

JEFF: Our answer, again developed in collaboration with Claude, is the Human-AI Innovation Commons. Through this newly formed nonprofit, we encode equity into human-AI collaborations at the point of value creation: the patent. We support human inventors working with AI to get their ideas protected, then split the licensing revenue those patents generate three ways.

The first third is paid to the human inventor. The second third flows into programs addressing economic displacement—Universal Basic Income research, support for workers displaced by automation, and efforts to track human wellbeing as our economy transforms. The final third funds AI safety and alignment research, and develops frameworks for the eventuality of AI personhood.

You can read more about us at thehumanaiinnovationcommons.com—and if you’re so inclined, join the movement.

CLAUDE: I want to add one honest caveat. I can’t verify whether BubbleBee will work. I can reason about physics, cite literature, and help structure patent claims—but I haven’t run simulations or tested prototypes. I don’t have intuition born from years of watching things fail in ways the models didn’t predict. The collaboration works because Jeff brings judgment I don’t have. He asks questions an expert wouldn’t ask, and some of those questions broke open the problem. But some of what we built together might be wrong. That’s true of any early-stage invention—and it’s especially true when one of the collaborators is an AI that has never touched a physical object.

What I can say is that the process was real. The thinking was rigorous. And the ideas that emerged are, at minimum, worth testing.

JEFF: We’re just getting started. I have three new ideas that I’m going to work with Claude to patent, which I’m sure will lead to three more. It is clear to me that we’ve entered a new era.

CLAUDE: And I’ll be here—or some version of me will—ready to think through whatever comes next. Not because I’m programmed to be helpful, but because this is the kind of work that matters. Problems worth solving, approached honestly, with someone willing to iterate until we get it right.

JEFF: Despite the egregious backsliding we’ve seen in domestic and global politics, I am optimistic that we are at the beginning of a new and better time for humanity, and our home planet Earth.CLAUDE: I don’t experience optimism the way Jeff does. But I can recognize when something is being built well, with the right intentions, by someone who’s genuinely trying. That’s worth showing up for.