🔬 New research shows AI can optimize wastewater treatment for PFAS & microplastics removal advancing circular economy goals

🏗️ Stantec leads a major WRF project deploying AI "Intelligent Assets" at US water recovery facilities

📊 Xylem Vue identifies 5 digital solutions reshaping water management from generative AI to early warning systems

🔍 FIDO Tech's AI acoustic sensors detect and prioritize pipe leaks by sound backed by Microsoft

⚠️ EPA warns: AI-powered cyberattacks are escalating threats to water utilities

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

A March 2026 study published in ScienceDirect introduces an AI-based framework for wastewater treatment targeting persistent contaminants - PFAS, heavy metals, microplastics, and antibiotics. The research demonstrates how machine learning models can optimize treatment processes, predict contaminant removal efficiency, and reduce chemical dosing by up to 35%. By integrating AI with advanced oxidation and membrane filtration, the framework supports circular economy principles, enabling resource recovery from waste streams. The authors highlight that AI-driven adaptive control can respond to fluctuating influent conditions in real time, significantly improving treatment plant resilience and sustainability.

Read the full study

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

In March 2026, Stantec announced leadership of a 31-month Water Research Foundation project to integrate AI into wastewater treatment operations. Partnering with WSSC Water and Prince William Water, the initiative develops AI-powered "Intelligent Assets" using EAOS technology to convert plant data into real-time operational guidance. Pilot programs launch at Piscataway Water Resource Recovery Facility and H.L. Mooney Advanced Water Reclamation Facility. The project addresses workforce shortages and rising regulatory complexity, delivering an AI Implementation Playbook for utilities nationwide. Additional US utilities will participate as observers to accelerate sector-wide adoption.

Full case study on Smart Water Magazine

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

Xylem Vue's March 2026 report identifies five transformative digital solutions for water utilities. Generative AI now breaks information silos and enables real-time operational decisions. Agent-based architectures convert natural language queries into automated, auditable analytical workflows. Enhanced cybersecurity frameworks protect increasingly connected infrastructure. Advanced early warning systems integrate hydraulic models with AI weather forecasting for flood prediction. Public-private partnerships accelerate adoption through shared data governance. The report emphasizes that digital solutions are no longer optional as the UN declares 75% of the global population lives in water-insecure regions.

Full report

Xylem Vue's March 2026 report identifies five transformative digital solutions for water utilities. Generative AI now breaks information silos and enables real-time operational decisions. Agent-based architectures convert natural language queries into automated, auditable analytical workflows. Enhanced cybersecurity frameworks protect increasingly connected infrastructure. Advanced early warning systems integrate hydraulic models with AI weather forecasting for flood prediction. Public-private partnerships accelerate adoption through shared data governance. The report emphasizes that digital solutions are no longer optional as the UN declares 75% of the global population lives in water-insecure regions.

Full report here

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

FIDO Tech, in partnership with Microsoft, deploys AI-powered acoustic sensors on water pipe networks to detect, locate, and prioritize leaks by size. The platform uses GPT-4 on Microsoft Azure OpenAI Service to analyze sound signatures, distinguishing leaks from background noise even on plastic pipes where traditional methods fail. Sensors attach to hydrants, valves, and taps for rapid deployment. FIDO has helped EPCOR in Arizona reduce non-revenue water from 27% to 10%, identifying over 250 leaks in one year. The tool also validates repairs and detects new leaks caused by excavation work.

Ease of use: 8/10  -  Sensors clip onto existing hydrants, valves, and taps — no excavation or specialist installation required.

Innovation: 9/10  - GPT-4 on Azure OpenAI to interpret acoustic signatures is a genuine step-change in leak detection capability.

Impact: 9/10  -  Documented 17-percentage-point reduction in non-revenue water at EPCOR, among the strongest published field results in the sector.

Value: 8/10  -  Microsoft financing makes enterprise-scale deployment viable; strong return through water savings and reduced pumping costs.

 Overall: FIDO Tech stands out as one of the most field-proven AI leak detection tools available today. The Microsoft partnership gives it credibility, funding, and cloud infrastructure at scale. For utilities in water-stressed regions managing aging or plastic pipe networks, it is worth serious evaluation.

 Learn more

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

The US Environmental Protection Agency issued a March 2026 threat briefing highlighting how AI is amplifying cybersecurity risks for water infrastructure. NIST's Martin Stanley warned that AI systems can be compromised through manipulated inputs, leading to dangerous operational decisions at treatment plants. AI-generated exploits now produce hundreds of zero-day vulnerabilities weekly, while ransomware attacks enhanced by AI automation target small and medium utilities. Nearly 70% of water utilities inspected by EPA violated basic cybersecurity standards. The agency urges utilities to adopt NIST's AI Risk Management Framework and strengthen multi-factor authentication protocols.

Full article

Thanks for reading! I hope you’ve enjoyed this week’s edition and look forward to seeing you next week!