Three years ago, people asked “is this written by GPT?” Today, they’re asking GPT whether it’s safe to eat a rotten avocado.
That shift tells you everything about where we are.
I grew up in the dot-com era. Watching technology spread back then felt slow – a phone would appear in one household, then another, then years later become universal. GPT hit a billion users in roughly three years. By any measure, artificial intelligence has normalized. The question now is what we do with that.
Understanding what’s actually at stake
The AI tools most people use, ChatGPT, Claude, Gemini, are built to follow instructions, learn from context, and complete specific tasks. They’re powerful, but bounded. AGI (Artificial General Intelligence) is a different category: a theoretical system that could match or exceed human cognitive ability across every domain, not just the tasks it was trained for.
Companies like DeepMind have publicly committed to pursuing AGI. Whether that’s achievable is debated — there are well-funded examples of teams spending years and enormous budgets without meaningful progress. It’s possible AGI remains theoretical indefinitely.
What isn’t theoretical is the world we’re already in.
The normalization nobody voted on
Whether you’ve adopted AI tools or not, you’re operating in an environment shaped by them. The content you read, the products you use, the workflows your competitors run: AI is embedded in most of it now. Opting out isn’t really an option.
(I used to publish three articles a week. I don’t anymore — it felt increasingly difficult to justify the effort in a content landscape this saturated. This is one of the first I’ve written in a while.)
What this means for business
At Albusi, we’ve been in business intelligence for 12 years — financial models, investment documents, pitch decks for clients across Europe and beyond. A detailed financial model once meant 2–3 weeks of work and fees around $5K. Today, a well-prompted AI tool can produce a credible first draft for a fraction of that.
We adapted. We updated how we work, how we price, and what we offer. And in doing so, we built something to help others do the same.
An agricultural client deciding whether to launch a seed product in a specific market window. A fintech company that needs a financial model this week, not next month. These aren’t edge cases — they’re the norm now. The gap we kept seeing was between raw AI output (fast but unreliable) and traditional professional work (reliable but slow). That gap is where Finixr fits.
AI normalization isn’t coming. It’s here. The organizations that treat that as a threat will keep losing ground to the ones that treat it as infrastructure.