The LK-99 superconductor and the lesson of hype

In 2023 a technical name jumped from the labs to the global trends: LK-99. A group of researchers presented what they described as a superconductor capable of working at room temperature, something that, if true, would change energy, transportation, and electronics. Within days the topic went viral: videos of “levitating” rocks, euphoric threads, and headlines talking about an imminent revolution. The enthusiasm ran far faster than the evidence.

Then came the boring but important part: other teams tried to reproduce the experiment, measured carefully, and compared results. Little by little, the extraordinary explanation deflated and more ordinary causes remained for what had been observed. LK-99 didn’t change the world, but it left a lesson that’s genuinely useful for any company: viral isn’t the same as true, and confusing them is expensive. At Normandia Web we see that tension every day when a small business wants to “jump on” the latest technology before knowing whether it suits them.

Hype isn’t a business plan

Every so often a tool or a trend appears that “everyone” is adopting. The pressure not to fall behind pushes you to buy fast, hire without comparing, or migrate whole processes because of a flashy demo. The problem isn’t the novelty: it’s deciding with the emotion of the moment instead of with reliable information.

The LK-99 case shows the full cycle. First, an enormous promise. Then, amplification on social media. And finally, the verification that separates the real from the apparent. Your company lives smaller versions of that cycle all the time, and whoever learns to wait for the third stage makes better decisions.

Viral grabs your attention; verification protects your money, your data, and your time.

What to check before believing (and before buying)

Before adopting any technology or tool, it’s worth cooling the fever with concrete questions:

• Who’s claiming it, and what do they gain from you believing it? A result verified by third parties isn’t the same as a promise from whoever’s selling.
• Can it be reproduced in your context? A perfect demo with ideal data doesn’t guarantee it works with your processes and your real operation.
• What problem of yours does it solve, specifically? If you can’t name the pain it removes, you’re probably buying a fad, not a solution.
• What happens to your data and your history? Every new tool should keep your data, processes, and history, not force you to start from scratch.
• How is the result measured? If there’s no way to know whether it worked, there’s no way to know whether it was worth it.

Critical thinking applied to your operation

Verifying before believing isn’t distrust; it’s method. In practice it means swapping “this looks amazing” for “let’s try this on a small scale and measure what happens.” A well-defined pilot gives you your own evidence: how it responds with your customers, with your team, and with your real volumes, not with a laboratory scenario.

That same logic is what we apply when we build custom software. Instead of imposing a generic tool that forces you to redo the way you work, we start from your real process and automate what actually moves the needle. That way technology adapts to your business, not the other way around. Artificial intelligence fits perfectly into this approach: it’s very useful when it solves a concrete, measurable task, and just noise when it’s adopted “because it’s trendy.”

How to evaluate the next novelty

If you want to take advantage of technology without falling for the hype, a sensible path is this:

• Name the problem first. Write in one sentence what you want to solve before looking at tools. The problem leads; technology follows.
• Start narrow and measurable. Choose a small process, define how you’ll measure it, and test for a short period.
• Demand evidence, not promises. Ask for real examples, references, and a test in your own context before committing budget.
• Protect continuity. Make sure you keep your data, processes, and history; no novelty should cost you your operational memory.
• Decide with reliable information. Compare results, not headlines, and scale only what has already proven it works.

The LK-99 story cooled off in weeks, but the lesson still stands: the best decisions aren’t made at the peak of enthusiasm, but after verifying. If you’re interested in applying technology with that cool head—automation, AI, and software built to your measure—at Normandia Web we help you start with what really matters: your problem, your data, and a result you can measure.