PFAS in Consumer Products: How OFF-PFAS Will Improve Risk Assessment
An international research project connecting data, artificial intelligence, analytical chemistry and policy
OFF-PFAS Consortium | Project overview
Per- and polyfluoroalkyl substances, known as PFAS, are used in many products and can persist in the environment. Yet monitoring and available information cover only a small part of the wider PFAS picture.
OFF-PFAS is a 48-month Horizon Europe project that brings together eight organisations from Bulgaria, Greece, Norway, Türkiye and Senegal. The project will combine research, industrial experience and staff exchanges to improve how PFAS in consumer products and the environment are identified and assessed
Key Insights
- PFAS are a large and diverse group of synthetic substances. Different PFAS can behave differently, so researchers need information on individual compounds and how people and the environment may encounter them.
- Many current monitoring approaches focus on a limited number of known PFAS and may miss precursors, sometimes called “hidden PFAS”.
- OFF-PFAS will connect a public PFAS information platform with AI and machine-learning tools, improved analytical methods and sampling in five countries.
- The project aims to provide evidence for researchers, policymakers and industry, while improving public understanding of PFAS.
- Planned project targets include information on at least 2,500 PFAS, four training schools and public engagement reaching at least 50,000 citizens.
Why a fuller picture of PFAS is needed
More than 10,000 PFAS have been manufactured since the 1940s. Many have been used in applications such as food packaging, textiles, cosmetics, nutraceuticals and industrial processes. Many PFAS are highly persistent and can be found in environmental media such as water, soil, air and dust.
The number of PFAS is far greater than the number routinely monitored. Information is also spread across separate databases and research fields. This makes it difficult to connect what is known about a substance with where it occurs, how it may affect people or the environment, and how it behaves over time.
One important gap concerns PFAS precursors. These substances can transform into other PFAS through chemical or environmental processes. If monitoring looks only for a predefined list of compounds, it may not capture the full PFAS content of a sample.
OFF-PFAS through the STAR framework
Situation: monitoring captures only part of the picture
OFF-PFAS starts from a substantial evidence gap. The project materials report that many current monitoring approaches cover fewer than 50 compounds, while more than 10,000 PFAS have been manufactured. Toxicological information is also limited for much of this wider group.
A single test or database cannot answer every question. Researchers need to connect data about chemical identity, toxicity, product occurrence, environmental presence, regulation and environmental fate.
Task: build an integrated basis for PFAS risk assessment
The project’s task is to improve the evidence available for assessing PFAS in consumer products and industrial settings. It will examine both targeted PFAS, which are already known and included in a test, and hidden PFAS that standard targeted analysis may miss.
The project will also develop ways to make research data more accessible and to use modelling to identify patterns, predict potential occurrence and explore possible long-term environmental outcomes.
Action: connect six work packages
OFF-PFAS has organised its work across six connected work packages.
WP1 will collect and organise PFAS information. The team plans to compile data on at least 2,500 PFAS and create an open-access information platform. It will include search and visualisation tools, a consumer-friendly information area and a module tracking relevant legislation.
WP2 will develop AI and machine-learning tools. These tools will help predict PFAS occurrence, estimate toxicity and model how PFAS may move and accumulate in the environment over five to ten years. The models will support research and decision-making, while laboratory work will remain essential for validating predictions.
WP3 will improve analytical methods. Researchers will harmonise methods for known PFAS and develop approaches to identify hidden PFAS, including precursors that can be transformed into compounds that are easier to measure.
WP4 will assess samples and risks. The project will collect consumer products, industrial materials and wastewater across five countries. It will also study manufacturing processes with industry partners, then compare analytical findings with relevant thresholds and assess how hidden PFAS may affect current risk estimates.
WP5 will share findings and support their use. Planned activities include open-access publications, conference presentations, policy briefs, an advisory document, industry engagement and multilingual public information.
WP6 will coordinate the project and develop skills. It will support 257 person-months of staff exchanges across the consortium and organise at least four training schools covering PFAS analysis, AI and machine learning, science-policy work and transferable skills.
Result: useful evidence, methods and public information
The project aims to produce an open PFAS information platform, predictive tools, validated analytical methods and a risk assessment based on sampling across five countries. It will share evidence with policy and regulatory audiences and develop practical information for industry and consumers.
Its public engagement target is to reach at least 50,000 citizens. This includes a planned PFAS Consumer Hub with an interactive product checker, factsheets, videos, social media content and public events.
These are planned project outputs and targets. Their purpose is to strengthen the evidence available for future research and decisions, not to provide a certification of individual products.
Key concepts explained
PFAS stands for per- and polyfluoroalkyl substances. It describes a broad family of synthetic substances with different chemical and physical properties.
Targeted analysis tests for a selected list of known compounds. It can provide detailed results for those targets but may overlook substances outside the list.
Non-targeted analysis uses analytical data to search more broadly for chemical signals, including compounds that were not selected in advance. Further work may be needed to identify and confirm them.
Hidden PFAS is a project term for precursors, contaminants or other PFAS that may not be captured by routine targeted testing. Some precursors can transform into other PFAS over time or under certain conditions.
AI and machine learning are computational methods that can identify patterns in data and generate predictions. In OFF-PFAS, models will help estimate occurrence, toxicity and environmental fate. Their outputs need to be interpreted and tested against scientific evidence.
Risk assessment considers what a substance is, where it occurs, how exposure may happen and what is known about its effects. Detecting a substance in a sample does not, on its own, determine the level of risk.
About OFF-PFAS
OFF-PFAS is a Horizon Europe MSCA Staff Exchanges project running for 48 months. It brings together eight academic and non-academic organisations from Bulgaria, Greece, Norway, Türkiye and Senegal. The project began on 1 November 2026 and is funded under Grant Agreement No. 101344717.
Frequently Asked Questions
Does OFF-PFAS study every PFAS?
No. The project aims to compile information on at least 2,500 PFAS and will use selected sampling and analytical approaches. Its goal is to improve the evidence base, including information on compounds that current monitoring may miss.
Why can standard tests miss hidden PFAS?
Many targeted tests look for specific known compounds. A precursor may not appear in the results unless the method is designed to detect it or transform it into a measurable form.
Will AI replace laboratory testing?
No. AI and machine learning can help identify patterns and guide research priorities. Laboratory methods are needed to measure compounds and validate model predictions.
Does detecting PFAS mean a product is unsafe?
Not by itself. Risk assessment needs to consider the specific substance, its concentration, how exposure may occur and the evidence about its effects.
Who can use the project’s results?
Researchers, policymakers and regulators, industry and the public. The project plans both scientific and policy outputs, as well as accessible information for consumers.