Water is a conservative sector for good reasons: assets last a century, failures are public health events, and the regulatory environment rewards reliability over novelty. Innovation that has actually landed tends to be unglamorous.

What has genuinely changed

Energy recovery in desalination
Pressure exchangers recovering energy from the concentrate stream took seawater reverse osmosis from around 8 kWh/m³ to 3–4. This is the single largest efficiency gain in modern water technology.
District metering and acoustic leak detection
Dividing networks into measurable zones and listening for leaks transformed leakage management from reactive repair to systematic detection.
Smart metering
Frequent automated readings identify continuous flow that indicates a leak on a customer’s own pipe — often the largest single loss a household has and otherwise invisible.
Membrane bioreactors
Combining biological treatment with membrane separation produces reuse-quality effluent in a fraction of the footprint, which is what makes urban water reuse practical.
Online water quality sensing
Continuous turbidity, chlorine and pressure monitoring detects failures in minutes rather than in the days a microbiological result takes.
Satellite leak detection
Radar satellite analysis to identify likely leak areas at network scale, narrowing where ground crews search.

Digital twins

A hydraulic model of a network, calibrated against live sensor data, allows operators to test interventions before making them and to detect anomalies against expected behaviour. The concept is genuinely useful. The limiting factor is almost never the modelling; it is asset data quality. Many utilities do not know precisely where their pipes are, what they are made of, or when they were laid, and a digital twin of an inaccurately recorded network is an accurate model of the wrong thing.

What is oversold

Misleading in context
Atmospheric water generation can solve water scarcity.

Condensing water from air works, and is genuinely useful for small-scale emergency and remote applications. But it is thermodynamically expensive — energy demand rises steeply as humidity falls, which means it performs worst exactly where water is scarcest. Per cubic metre it is far more energy intensive than desalination. It is a niche technology being marketed as a general solution.

Uncertain
Graphene membranes will make desalination cheap.

Laboratory results for graphene oxide and other advanced membranes are genuinely impressive on permeability. But desalination energy use is dominated by the thermodynamic minimum work of separation, not by membrane resistance — modern plants already operate within a factor of about three of the physical floor. A better membrane reduces pumping losses; it cannot go below the thermodynamic limit. The achievable gain is real but incremental.

Sources