America set its first PFAS drinking water limits. Two years on, it is unpicking some of them.

For decades the United States had no federal limit at all on PFAS in drinking water. States acted individually or not at all, most water systems were not required to test, and the result was that nobody could say with confidence how widespread the contamination was — while research kept finding the compounds in the blood of very nearly everyone tested.

That changed on 10 April 2024, when the EPA finalised the first national drinking water regulation for PFAS. It set enforceable limits of 4 parts per trillion for PFOA and for PFOS — extraordinarily low numbers, and deliberately so — along with limits of 10 parts per trillion for PFHxS, PFNA and GenX chemicals, and a combined hazard index covering mixtures of those plus PFBS. Public water systems were given until April 2029 to comply.

Two years later that framework is being reshaped, and anyone relying on it should know exactly which parts still stand.

In May 2026 the EPA proposed to keep the maximum contaminant levels for PFOA and PFOS at 4 parts per trillion, while allowing water systems to request two additional years — to 2031 — to meet them. A separate proposal would rescind the limits for PFHxS, PFNA and GenX, and the hazard index along with them. The comment period on the extension closed in July 2026.

So the practical position today: the two most-studied compounds remain regulated at a very strict level, with the compliance date moving out; the limits on the newer replacement compounds are proposed for removal at federal level. A number of states keep their own standards regardless of what happens federally, which means the answer to "is this water regulated for PFAS" increasingly depends on which state you are standing in.

It is worth putting that beside what Europe did. The EU’s recast Drinking Water Directive brought its own PFAS limits into application on 12 January 2026 — the sum of twenty specified compounds at 0.10 micrograms per litre, or total PFAS at 0.50. The two systems are built differently: the American approach names a few compounds and regulates each tightly, the European one sums a broad group at a higher combined threshold. Neither covers a family that runs to thousands of compounds, and both share the same weakness, which is that the replacement chemicals arriving on the market are ahead of the regulation of the ones they replaced.

The short-chain PFAS that industry moved to are also, inconveniently, the hardest to filter. Being smaller molecules, they pass through treatment that captures their long-chain predecessors. A regulation that names PFOA and PFOS does not follow the chemistry when the chemistry moves.

None of which is an argument for despair, and it is certainly not an argument against regulation — a 4 parts-per-trillion federal limit on PFOA is a serious public health measure that did not exist three years ago. It is an argument for not mistaking a rule for a result. Between a limit being set and a contaminant leaving a particular tap sits testing, funding, treatment plants, construction, and years.

Our concern is narrower than national policy and more immediate. The children in the homes and shelters we serve are drinking water this week, from a building none of them chose, and neither a compliance deadline in 2031 nor one in 2029 changes what comes out of that tap tomorrow morning. What changes it is a system at the point of use, chosen against a test of what is actually in that building’s water, and maintained afterwards.

That is why we ask for proof rather than specifications. Our partner NearonYdor had its Twin system tested for fifteen PFAS compounds by LC-MS/MS at the Chemistry Department of the National and Kapodistrian University of Athens — deliberately including the short-chain compounds that most filters fail on. After filtration every one of the fifteen fell below the limit of detection. We would rather fund equipment somebody independent has tried to break.

Sources: US EPA, PFAS National Primary Drinking Water Regulation, 10 April 2024; EPA news release "EPA Announces It Will Keep Maximum Contaminant Levels for PFOA, PFOS" and the proposed compliance extension and rescission rules, May 2026; Directive (EU) 2020/2184; laboratory report, National and Kapodistrian University of Athens (Chemistry Department).