While we've been evaluating AI-enabled businesses and discussing the transformative effect these tools are currently having and are capable of providing, something important was staring us in the face: talking about, reading about, and conversing with AI is not the same as leveraging it.
There's a critical distinction between AI Literacy and AI Fluency. One gets you here, the other takes you there.
AI literacy means understanding what AI can do and engaging in informed conversations about its potential. AI fluency means actually picking up the tools, getting your hands dirty, and developing an intuitive sense of how to leverage AI to solve real problems.
Most professionals today are stuck in the literacy phase. We know Claude can write and GitHub Copilot can code, but knowing about these capabilities and actually developing workflows that enhance our productivity are worlds apart.
The risk of staying in literacy mode? You become the person who "understands" AI but can't actually use it when it matters.
Rather than scheduling another meeting about AI adoption, we blocked out two hours for a team work session. The rules were simple:
Right away the team got to work picking up some tools they’ve used before and some they hadn’t. A few team members tackled market research & GTM strategies. Using Perplexity for market research, Claude for data analysis, Lovable for landing pages, and Relevance to build "Ollie," an automated outreach agent. A go-to-market strategy in two hours.
Another solved a persistent customer Q&A documentation problem by prototyping an AI system to convert Slack conversations into polished, categorized content for public use. Other work included creating a streamlined application for capturing financial files to accelerate our LOI process, an AI agent for initial investment research, a deployment assistant that could generate infrastructure code from plain English instructions.
The Prompt Learning Curve Is Real: Everyone struggled initially with getting AI tools to understand exactly what they wanted. But within 30 minutes, intuitive prompting strategies emerged through trial and error.
Tool Integration Matters: Our GTM project demonstrated how AI tools work best in combination—orchestrating multiple tools to achieve results no single tool could deliver alone. This systems-level thinking emerges once limits are tested in individual tools.
Real Business Impact Is Possible: These weren't toy projects. Each solution addressed actual business challenges—expansion strategy, internal operations, investment workflows, customer service, and technical infrastructure. The time constraint forced creativity and got us away from perfect being the enemy of good.
Our hackathon was a microcosm of a larger shift happening across industries. The professionals who will thrive won't necessarily be those who understand AI the best—they'll be those who can most effectively integrate AI into their daily workflows.
This integration skill can't be learned through observation. You can watch a thousand YouTube videos about prompt engineering, but until you've spent time refining prompts for your specific use cases, you won't develop the intuitive sense of how to communicate effectively with AI systems.
These are early steps in a path to AI Native working – skill building in prompt engineering, tool integration, quick prototyping, and collaboration/peer review is the foundation of being able to utilize AI as an ingrained part of working.
The Compound Effect
What makes AI fluency particularly powerful is its compound nature. Each workflow improvement builds on the last.
The market research workflow immediately suggested applications for other portfolio companies. The documentation automation sparked ideas for AI-powered knowledge management. The LOI application opened possibilities for automating other due diligence processes.
These compound improvements don't happen overnight, and they don't happen through passive consumption of AI content. They happen through consistent, hands-on practice with real problems and real constraints.
Our 2-hour hackathon was just the beginning. We're building regular "AI sprints" into our team rhythm—short, focused sessions where we tackle real problems with AI tools.
The approach is straightforward:
Start with problems that matter to you personally. When the problem is real, you'll push through the initial learning curve.
Give yourself permission to build imperfect solutions. The goal is learning, not shipping production-ready code.
Set tight time constraints. Artificial urgency forces you to focus on what actually works rather than getting lost in possibilities.
Share your experiments with colleagues. Learning compounds when it's social. The show & tell afterward allows for discussion and identifying ideas that can be pushed further.
The question isn't whether AI will transform how we work, but whether we'll be active participants in that transformation or passive observers of it.
Becoming AI-native isn't about mastering every tool or understanding every technical detail. It's about developing the confidence and intuition to reach for AI solutions when real problems arise.
Our hackathon taught us that the gap between AI literacy and AI fluency isn't as wide as we thought—but it does require real work and real time spent.
We’re excited to see the fruits of this labor shape the way Curious and its companies do business, as well as what we can continue to learn in the future from others doing the same and those who are exploring this work in different ways than we are.
Happy building.🧑💻
Curious is a long-term holding company that buys and grows software companies with empathy.
Join the Curious Dispatch to receive our latest thinking, news, and opportunities within our portfolio.
