It is a fair question, and one of the first that thoughtful people ask before they recommend exora to someone they care about: what about AI and climate change? The energy. The water. The headlines.
You deserve a straight answer, not a brush-off. So here it is.
The short version: the energy and water cost of using exora is real but small, and quite different from the headline AI numbers most people have in mind.
The scary part of AI is building the models. We do not do that.
The frightening figures you have read - data centres the size of suburbs, huge amounts of cooling water - mostly come from building the AI models in the first place. That is called training, and it is the resource-hungry part. It happens once, up front, by the big companies that make the models.
We do not build those models. We use finished ones, and only in short bursts - a few seconds of work when you upload a document or ask a question, and then nothing. There is no model running in the background, burning power while you sleep. It works when you ask it to, and then it stops.
Using a finished model still costs something, and I want to be honest about that. Because billions of people now do it every day, it does add up across the whole world. But our share of that is small: one person’s documents, a few seconds at a time.
The work is careful, not wasteful.
For most of the work we deliberately use smaller, efficient models, and only reach for the bigger, hungrier ones for the harder, more complex steps. Every answer matters - but not every step needs the heaviest tool to get it right. That is partly about cost and partly about quality, but it has the same effect: less computation for the same result.
To put it in everyday terms: answering one question uses about the same amount of electricity as a single internet search. Processing a whole document is more like a handful of searches. That is the honest scale of it - closer to an ordinary minute online than to anything dramatic.
Where it runs matters too.
Our processing runs in Sydney, mainly so Australian health data stays in Australia. A nice side effect is that it sits on the Australian electricity grid, which is getting cleaner every year as the country shifts to renewable energy. As the grid gets cleaner, so does every document exora processes - without you having to do anything.
What about the water?
Water is the other worry in the headlines, and it is a fair one. AI uses water in two main ways: to cool the computers inside a data centre, and to generate the electricity those computers run on. As with energy, the amount tied to one person’s documents is tiny, and honest estimates vary a lot depending on how you measure it.
There is a quiet bit of good news here too. Wind and solar use far less water to produce power than coal and gas do. So as the grid shifts to renewables, the water hidden inside that electricity quietly shrinks as well.
And here is the part everyone forgets.
The AI-and-climate conversation almost always leaves out the other side of the ledger: healthcare itself is one of the most resource-hungry things we do. If the world’s health systems were a country, they would be the fifth-largest polluter on the planet - around 4 percent of all global emissions. In Australia, healthcare is responsible for about 7 percent of the nation’s carbon emissions.
And a lot of that is wasted. Researchers estimate that roughly a third of healthcare is low-value - tests, scans and treatments that do not actually help. It is not an abstract problem. In Australia alone there are about 80 million community blood tests a year, and an estimated 10 to 40 percent of them are likely unnecessary - often because one clinic simply could not see what another had already ordered. Every one of those repeated tests, duplicate scans, and extra trips across town burns energy, water and carbon.
This is exactly the gap exora is built to close. When your health data is unified in one place - and you can walk into an appointment with it - that can mean one fewer repeated test, or one better-informed visit. A single avoided scan or repeated round of blood tests saves far more energy, water and carbon than the AI ever used to help you avoid it.
The honest bottom line.
Yes, there is a cost. I will not pretend there is not. But it is small, it happens only when you use it rather than running constantly, and for most people it is paid back many times over through less wasted care - fewer appointments, fewer repeated tests, fewer trips.
We would rather give you the real picture than a comfortable one. That is the same principle behind everything else we build: exora shows you where every fact came from, and we would rather tell you the honest answer here too.
A note on the numbers: per-use AI energy and water figures vary a lot depending on who is measuring and how, so we have kept to honest ranges rather than false precision. Sources are linked throughout, and figures reflect independent analysis as of mid-2026. We will update this post if the picture changes.
