The Lonely Work of Innovation: Why Being Right Too Early Still Beats Being Late

I recently interviewed Thomas Altman on the Sidecar Sync podcast, and his story perfectly illustrates the innovation journey many of you are already on. 

Thomas had been watching AI developments since 2020, and by 2022 he and his co-founder Dray McFarlane teamed up with Blue Cypress and founded Tasio (now TasioLabs) to bring AI solutions to associations. The problem? Nobody was taking AI seriously yet. They were experimenting with GPT—not ChatGPT, but its predecessor that was essentially sophisticated autocomplete.

Thomas describes those early days as screaming in the wind about AI's potential while others dismissed it. Sound familiar? Maybe you're the one in your organization talking about AI agents, world models, or some other emerging technology that makes people's eyes glaze over.

But here's the payoff: When ChatGPT launched in late 2022, his team launched Betty, an AI knowledge assistant for associations, just a few months later. While other organizations scrambled to understand what had happened, they were ready to ship products.

The lesson? Being prepared for emerging tech means laying track before the train arrives.

The Early Adopter's Reality Check

Thomas saw something others often missed: associations were sitting on massive datasets with untapped potential. Member data, transaction histories, engagement patterns—all collecting digital dust while associations struggled with member retention and engagement.

The challenge wasn't just technical. It was solving real problems with technology that seemed primitive. Imagine trying to convince your board in 2021 that autocomplete technology would revolutionize member services. You'd get the same polite but skeptical looks Thomas got.

During our conversation, Thomas was matter-of-fact about it: nobody takes you seriously when you're years ahead of the curve. But that's not a bug in the innovation process—it's a feature. If everyone immediately sees the value in what you're seeing, you're probably not looking far enough ahead.

The "Laying Track" Strategy

For years before ChatGPT became a household name, Thomas and his team were developing expertise in generative AI. They weren't sure exactly how it would transform associations, but they knew it would.

They experimented with the member retention problem—a challenge every association faces. Could predictive modeling help? What about personalization at scale? They tried different approaches, hit dead ends, pivoted, and kept building their understanding of what these models could do.

When ChatGPT exploded onto the scene, Thomas's team didn't need to scramble to understand the technology. They'd been living with it for years. Within a few months, they transformed their accumulated expertise into Betty, solving the knowledge discovery problem that plagues every association with a substantial content library.

The track they'd been laying suddenly had a train on it.

Finding Your Fascination Point

Thomas shared advice during our interview that cuts through all the noise about innovation:

Lean into what fascinates you.

This isn't feel-good career advice. It's practical strategy. If something captures your attention—whether it's AI agents, blockchain, quantum computing, or something else entirely—it's probably for a reason. Your instincts are picking up on potential that your rational mind can't quite articulate yet.

Thomas couldn't have predicted in 2020 exactly when or how AI would transform associations. But his fascination with AI and association data led him to build expertise that became invaluable at exactly the right moment.

You can't predict the exact future. None of us can. But following your curiosity prepares you for futures you can't yet imagine.

Building Credibility Through Small Experiments

The approach depends on where the technology is in its lifecycle.

When tools exist (like ChatGPT today): Start with small bets that require minimal time investment. Run an experiment this afternoon. Document what worked. Share it casually with your team. Show real, productive value—automating a tedious task, analyzing member feedback in a new way, creating a first draft that usually takes hours. Each small win builds credibility.

When the technology isn't accessible yet (like world models today): You can't demo what doesn't exist, but you can still build credibility. Document the problems it would solve. Share articles and research. Run thought experiments: "What if we could simulate our entire conference before booking venues?" Connect future possibilities to current pain points.

Thomas faced both situations. Early on, he couldn't show GPT doing much—it was just fancy autocomplete. So he focused on the problems: member retention, knowledge discovery, engagement challenges. He kept learning, kept talking about the potential, kept connecting the dots between emerging tech and association needs.

The key is staying engaged with the trajectory, not just the current state. When you can experiment, do it. When you can't, become the person who understands what's coming and why it matters. Either way, you're building the credibility you'll need when the technology becomes real.

By the time the technology goes mainstream, you'll be positioned as the person who saw it coming—whether you could touch it early or not.

Managing the Day Job While Building Tomorrow

How do you keep the lights on while preparing for the future?

The answer is refreshingly practical: dedicate 30-60 minutes a few times a week to staying in the conversation. Listen to podcasts during your commute. Join online communities during lunch. Experiment with new tools in the margins of your day.

You don't need to abandon your responsibilities to be an innovator. In fact, your day-to-day work is where you'll find the problems that emerging technology might solve. The key is maintaining that dual awareness—executing today's priorities while watching for tomorrow's opportunities.

Make innovation part of your existing role, not a separate initiative. When you're working on member engagement, ask yourself how AI might enhance it. When you're planning an event, consider how simulation technology might optimize it. When you're creating educational content, imagine how it might become interactive.

Practical Tactics from the Trenches

After talking with Thomas and reflecting on his journey, here are the tactics that actually work:

Accept the skepticism timeline. People won't take you seriously at first, and that's fine. You need that time to build expertise anyway.

Find your people. Even one or two allies who see what you see can sustain you through the early period. Thomas had his co-founder Dray. Who's yours?

Frame everything in member value. Don't talk about how cool the technology is. Talk about how it solves member problems.

Document everything. Your small experiments today become your proof points tomorrow.

Build expertise before you need it. The time to learn about world models isn't when they're mainstream—it's now, when you have space to experiment.

Stay ready to move. When the moment arrives—and it will arrive suddenly—you need to be able to act fast.

The Competitive Advantage of Being Early

Thomas's journey from those early days watching AI develop to launching Betty proves something important: the experimental work you're doing today will be someone's competitive advantage tomorrow. The question is whether that someone will be you.

Today, Thomas is exploring world models, AI agents living thousands of simulated lifetimes, and other concepts that might sound like science fiction. But he's learned to trust the pattern. Today's "impossible" is tomorrow's "obvious."

So what track should you be laying now? What fascinates you that others dismiss? What problems do you see that others ignore? What experiments could you run that cost almost nothing but might teach you everything?

The next wave of innovation is already building. World models, quantum computing, technologies we haven't even named yet—they're all coming. The associations that will lead that wave are the ones whose people are experimenting today, building expertise that may seem useless until suddenly it isn't.

You might be the "crazy AI person" in your organization right now. And the truth is... maybe that's exactly where you should be.