Inside aquAI - 27 Oct 2025 | Edition No. 15

Flow time: 5 min I your weekly pulse on AI news, tool and case studies reshaping the water sector

💧 AI predicts river flow – Cognizant’s open-source model uses weather and land data to forecast daily river flow, helping plan for droughts and pollution control.

🤖 Copilot comes to Edge – Microsoft adds AI directly into its browser, letting users summarise pages and compare data without leaving a tab.

🔬 AI detects microplastics – Clarkson University’s computer-vision system tracks how microplastics sink in water, cutting costs and improving clean-up planning.

🔧 Tool of the week: Napkin AI – Instantly turn text into flowcharts and mind maps - fast, simple, and perfect for reports and ideas.

🕵️ The Shadow of AI – A major AWS outage took Alexa, Canva, and others offline, exposing how dependent AI systems are on single clouds, resilience is key.

If you’re a decision maker at your company, you need to be on the bleeding edge of, well, everything. But before you go signing up for seminars, conferences, lunch ‘n learns, and all that jazz, just know there’s a far better (and simpler) way: Subscribing to The Deep View.

This daily newsletter condenses everything you need to know about the latest and greatest AI developments into a 5-minute read. Squeeze it into your morning coffee break and before you know it, you’ll be an expert too.

Subscribe right here. It’s totally free, wildly informative, and trusted by 600,000+ readers at Google, Meta, Microsoft, and beyond.

Cognizant’s River Deep Mountain AI team has shared an open-source machine-learning model that predicts daily river flow, even in places without gauges. The model has two parts: an LSTM branch that learns from changing weather over time, and a second branch that reads fixed catchment facts like soils, land cover, and elevation; the two are combined to make the final prediction. It was trained on 40+ years of data from 409 English catchments , the Environment Agency Hydrology Data Explorer, and Natural England chalk-stream layers, and the first release shows strong skill at many sites. Code and notebooks are available for anyone to test and adapt.

Accurate flow where gauges don’t exist helps plan drought responses, size treatment and abstraction, target pollution tracing, and cut monitoring costs. Because the model is open-source, utilities, regulators, and consultants can audit, tune, and embed it in planning tools, digital twins, and early-warning dashboards, speeding up decisions while keeping methods transparent.

👉 Full article

Microsoft introduced Copilot Mode in the Edge browser, an optional, AI-powered way to browse. It puts chat, search, and navigation in one place; can summarize pages, compare info across your open tabs, and even supports voice. If you opt in, Copilot can use the context of the page (and, in future updates, limited browser history/credentials for tasks like bookings) with clear on/off controls. It’s currently free for a limited time on Windows and macOS.

Having Microsoft Edge with Copilot built in means AI support is now part of your everyday browsing, no extra app or plug-in needed. It can summarize reports, compare technical data, and find answers instantly while you’re reading a webpage. For engineers, analysts, or project managers, this means less time switching between tabs or searching manually.

👉Full story

Researchers at Clarkson University built an open-source, AI-assisted computer-vision method that tracks how fast microplastic particles sink in water (their “settling velocity”). Using cameras plus machine learning, the system automatically follows each particle and calculates its speed with accuracy within ~6% of careful manual measurements. The team plans to extend this toward real-time, AI-powered detection, and their work is published open access in the journal Microplastics.

Knowing how quickly different plastics sink helps predict where microplastics collect -in rivers, reservoirs, and treatment plants. A fast, consistent, and sharable method means utilities and regulators can monitor more sites, prioritize clean-ups, and improve source control with better evidence. Because the workflow is open and automatable, it can lower costs and speed up monitoring programs, supporting clearer decisions on intake protection, sludge handling, and watershed management.

👉Full article

Napkin AI is a web app that turns your text into clean visuals, mind maps, flowcharts, and diagrams, to help you communicate ideas faster. You type or paste content, and Napkin generates editable visuals you can style and export.

🔌Try it

A major AWS (Amazon Web Services) outage on Oct 20, 2025 knocked many apps and devices offline, including Alexa, Snapchat, Canva, Perplexity, and more. The failure was traced to a technical issue in Amazon’s cloud infrastructure that affected one of its biggest regions (US-EAST-1). The incident reminded everyone just how dependent we’ve become on a few massive cloud systems that quietly power almost everything, including many AI tools and digital assistants.

This outage shows the hidden risk behind AI and cloud dependence. Most AI models from chatbots to water monitoring dashboards run on shared cloud platforms. If one fails, many connected systems fail too. For sectors like water, energy, and public infrastructure, this could mean lost data, delayed alerts, or downtime in essential services. The takeaway is simple: resilience matters. Organisations should design systems that can keep running when one cloud or AI service goes down — using backups, multi-cloud options, and clear offline plans. AI is powerful, but it’s only as reliable as the systems that support it.

👉Full story

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

Dr. Andrea G.T