Disinfecting drinking water produces potentially toxic byproducts — new AI model is helping to identify them

Hoboken, NJ., January 12, 2026 — Disinfecting drinking water prevents the spread of deadly waterborne diseases by killing infectious agents such as bacteria, viruses and parasites. Without disinfection, even clear-looking water can carry pathogens that can cause severe and even life-threatening illness, especially in children, older adults, and people with weakened immune systems. Before water disinfection processes were put in place, outbreaks of waterborne diseases such as cholera, typhoid, and dysentery routinely claimed lives, decimating cities and even countries. Disinfecting drinking water is one of the most important public health achievements in human history.

However, the chemicals commonly used to disinfect water, such as chlorine or chloramine, also react with organic matter, the tiny bits of dissolved organic carbon that is inherently present in water as it comes from natural sources such as rivers, lakes or aquifers. When these chemicals react with organic matter, they form byproducts that may be harmful to human health. Some of these disinfection byproducts, which scientists refer to as DBPs, have been implicated in certain cancers and reproductive issues. For example, DBPs like trihalomethanes and haloacetic acids have been linked to increased risks of bladder cancer as well as impaired fetal development.

The Environmental Protection Agency has standards for the safety levels of some of these byproducts present in the drinking water, but not for all, says Stevens Institute of Technology Assistant Professor Tao Ye, who uses AI to analyze environmental data to understand the complex interactions of various chemical compounds. “There are 11 such byproducts regulated by the EPA,” he explains. “However, so far research has identified several hundred more, which we don’t know much about — and they may be more toxic than the ones that are regulated.”

Although understanding how the chemistry of these compounds may affect human health is important, testing these compounds in laboratory conditions is challenging. “Traditional toxicity testing in the lab is often time-consuming, labor-intensive, and expensive, which limits how many disinfection byproducts can be evaluated,” Ye says. That’s where AI can help, he notes.  “AI and machine learning are fundamentally transforming this process by enabling rapid, scalable toxicity screening, allowing us to assess hundreds of compounds that would otherwise be impractical to test experimentally.”  

To speed up DBP research, Ye, his PhD student Rabbi Sikder, and their collaborator Peng Gao at Harvard T.H. Chan School of Public Health, built an AI model to help assess the disinfectant byproducts and their toxicity. 

First, researchers combed through scientific studies to find the toxicity data available from experimental testing of over 200 chemicals. Then, they trained the AI model on this data to predict potential toxicity for other chemicals. 

“We used the laboratory testing data reported in previous literature,” explains Sikder. “We collected those chemical names, their chemical structures, along with experimental exposure conditions and their corresponding toxicity values. We found toxicity values for 227 known chemicals and used them to build a machine learning predictive model to predict the toxicity for the unknown ones.”  

The model was able to predict the toxicity for 1163 byproducts of the cleaning process. The model also found that some of the byproducts’ potential toxicity ranged from 2 to 10 times higher than some of the chemicals that the EPA regulates, Sikder says. The team outlined their findings in the paper titled Multi-Endpoint Semi-Supervised Learning Identifies High- Priority Unregulated Disinfection Byproducts, published in the journal of the Environmental Science & Technology Letters on January 15, 2026. 

Does that mean that your tap water isn't safe to drink?  “Not at all,” Ye explains. Your average glass of tap water will never have all these harmful byproducts together. That is a total number of compounds that may theoretically form, depending on what organic matter is present in the water and what chemicals are used to clean it. In different parts of the world, water contains different organic matter and different chemicals are used to treat it, Ye stresses out. “What we are doing here is our due diligence to see what else may need to be regulated, depending on what’s in the water and what you use to clean it,” he says. “All in all, our tap water is safe to drink, and our research intends to make it even safer.” Sikder adds that now that the AI model exists and is available for use, other scientists can access it to further understand the chemistry of the DBPs.

For those who remain concerned, Ye shares his advice on how to remove the disinfectant byproducts from their tap water. “As researchers, we are always trying to do two things — advance the science and inform the public. The first thing in this case is understanding the mechanisms behind the formation of toxic compounds. And the second one is how to reduce these chemicals in our tap water, which you can do in two different ways. You can filter the water with various widely available household filters. Or you can boil it because when you boil it, these chemicals evaporate,” Ye says. “Both methods are easy to do at home.”
 

About Stevens Institute of Technology 

Stevens is a premier, private research university situated in Hoboken, New Jersey. Since our founding in 1870, technological innovation has been the hallmark of Stevens’ education and research. Within the university’s three schools and one college, more than 8,000 undergraduate and graduate students collaborate closely with faculty in an interdisciplinary, student-centric, entrepreneurial environment. Academic and research programs spanning business, computing, engineering, the arts and other disciplines actively advance the frontiers of science and leverage technology to confront our most pressing global challenges. The university continues to be consistently ranked among the nation’s leaders in career services, post-graduation salaries of alumni and return on tuition investment.