Most humidity sensors still work on logic that’s about 50 years old. They ask one simple question: Is humidity above 60% right now?

If yes: Alert.

If no: Safe.

That approach is simple — and scientifically wrong.

Mold doesn’t respond to a single number. It responds to temperature, humidity, surface material, and time. The same 60% RH may be harmless in one room yet dangerous in another. A two-hour spike after a shower is nothing like two weeks of dampness in a cold corner.

So instead of building another threshold sensor, we built AirGuard around building physics. Rather than asking “Is it humid?”, we ask: “Given the current conditions in this room, how long until mold will grow?”

The Science: From Thresholds to Timelines

This shift comes from three decades of research.

In the 1990s, researchers at VTT Finland developed a dynamic mold model. By exposing thousands of material samples to controlled conditions, they showed there is no flat “safe line.” Mold growth depends on the interaction of temperature and moisture over time.

Later, this framework was expanded to account for material sensitivity — drywall, for example, is far more vulnerable than concrete.

There was just one problem: these models assumed you had professional probes drilled into the walls. Homes don’t have that. They have air sensors.

The Breakthrough: The “Virtual Probe”

In 2022, researchers at the University of Exeter (Menneer et al.) adapted the VTT model for real homes using standard air measurements. They identified three keys to making air sensors accurately reflect surface risk.

1. Sensitivity. Air sensors systematically under-predict surface moisture because humidity dilutes across the room. To correct for this, the researchers introduced a new sensitivity class — “Very Sensitive ×2” — effectively doubling the parameters to match real household outcomes.
2. The 50% Rule. When driving the model from air data, the effective critical humidity is much lower — around 50% RH. Not because mold starts growing at 50%, but because cold surfaces can be near saturation while the room air still looks “normal.”
3. Responsiveness. The model was recalibrated to react faster to rising moisture, while still allowing risk to decline during dry periods.

Taken together, these changes turn an ordinary air sensor into a Virtual Surface Probe.

How AirGuard Works

AirGuard implements this Menneer-adapted VTT framework directly.

Instead of saying “humidity is high,” we compute a real-time Time-to-Growth forecast, modeling how temperature and humidity interact on surfaces under real-world conditions.

This allows us to classify risk as:

🟢 Safe: more than 90 days

🟠 Warning: 30–90 days

🔴 Critical: less than 30 days

We alert when the timeline crosses into risk — not when an arbitrary RH number flips. This enables earlier and more meaningful notification.

Physics + Chemistry: The Dual-Sensor Advantage

Predicting the future is powerful. But what if mold is already present, despite a current dry spell?

Physics helps us predict risk. Chemistry helps us confirm reality.

That’s why AirGuard pairs its forecast engine with gas-based detection.

Rather than forcing one chip to do everything, AirGuard uses a second, dedicated sensor tuned to detect mold-related VOC signatures. Once calibrated, this lets us distinguish between two very different situations:

Preventable risk: The air is damp, and the physics say growth is coming.

Hidden sources: The air looks dry (forecast safe), but the chemical sensor detects mold metabolites.

Two sensors. Two signals. Physics for prediction, chemistry for detection.

What’s Next: The Odometer for Moisture

Right now, our forecast works like a speedometer — it shows how fast you’re moving toward mold growth based on current conditions.

But biology has a memory. A room doesn’t instantly reset when you open a window, and it doesn’t forget weeks of dampness overnight.

That’s why we’re building the odometer: cumulative exposure tracking.

Future updates to AirGuard will track room history over weeks and months, accounting for prolonged moisture and recovery time. This will reveal when a room that looks dry is actually scarred by past dampness — and primed for rapid regrowth the next time conditions worsen.

AirGuard applies dynamic mold-growth modeling to standard air measurements in residential environments. This approach moves beyond fixed humidity thresholds to evaluate risk as a function of environmental conditions over time.

Such framing better reflects how mold actually develops in buildings.