What the New EPA PFAS Limits Mean for Your Water

In May 2026 the EPA proposed two changes to the EPA PFAS limits for drinking water, and the headlines made it sound like the limits were being scrapped. They are not. The 4 parts-per-trillion limits for PFOA and PFOS – the two PFAS your filter is most likely certified to remove – are being kept. What is changing is the deadline for utilities to comply – systems that apply could get two more years, to April 2031, while the extension is not automatic and systems that do not apply still face the April 2029 deadline – and the status of three other, less-common PFAS, whose limits the EPA has proposed to drop. The public comment window on both proposals was open through July 20, 2026. Here is the plain-English version of what changed, what it means for your tap, and how to weigh in before the window closes.

The honest headline: the limits that matter most for home filtration did not move. In our national PFAS study, about 1 in 9 US public water systems averaged PFOA or PFOS above the 4 ppt limit – a real, large number, but it also means most systems came in under it. The rule fight does not change the one useful move: check your own utility first.

What the new EPA PFAS limits actually say (two separate rules)

The confusion is because there are two proposals, not one.

1. Keep the limits, extend the deadline (the compliance-extension rule – docket EPA-HQ-OW-2025-1742). The EPA proposed to uphold the 4 ppt Maximum Contaminant Levels (MCLs) for PFOA and PFOS set in the 2024 rule, while giving water systems an option to request two more years to comply – from April 2029 to April 2031. The limits themselves do not change; the time utilities have to meet them does.

2. Rescind the limits for three other PFAS (the rescission rule – docket EPA-HQ-OW-2025-0654). The 2024 rule also set 10 ppt limits for three less-studied PFAS – PFHxS, PFNA, and GenX (HFPO-DA) – plus a “Hazard Index” for mixtures of them. The EPA proposed to rescind those, saying the proposal is based solely on correcting what it calls the unlawful procedure under which those limits were set. PFOA and PFOS are not part of the rescission.

Both proposals share the same public process: a virtual public hearing on July 7, 2026 and a written-comment deadline of July 20, 2026. (EPA: compliance-extension rule; EPA: rescission rule)

What it means for your water

Four honest takeaways, no panic:

• The filter math is unchanged. PFOA and PFOS at 4 ppt are the limits home filters are certified against. If your water is over that line, the fix at your tap is the same today as it was before the proposal.
• The clock on your utility is longer, not shorter. If your system reported PFAS over the limit, it may now have until 2031 to treat for it. That is the argument for handling your own tap in the meantime rather than waiting on the utility – the filter is the part you control.
• If your area’s PFAS is GenX, that is the one being dropped. The rescission targets PFHxS, PFNA, and GenX. Communities whose contamination is specifically GenX – the Cape Fear region of North Carolina is the best-known case – would lose the federal limit for it if the rescission is finalized. Worth knowing which PFAS your area actually has.
• Most systems still came in under the limit. Our study found about 1 in 9 systems over for PFOA/PFOS – which means roughly eight in nine were under. This is a check-yours situation, not a buy-a-filter-in-a-panic one.

A note on reading any PFAS number, including ours: EPA monitoring results are individual samples, and the legal limit is a running annual average, so a single result above 4 ppt is a flag to look closer, not a finding that a utility is “in violation.” We report what the federal record shows, with that caveat, and never make a health claim from a monitoring number. For the underlying health science – what the EPA and ATSDR actually conclude – see our PFAS contaminant guide.

How to check your own water

A national rule is not an answer about your house. The per-system data is:

1. Enter your ZIP in the water watchdog. It shows what your utility reported under the EPA’s UCMR5 program (2023-2025), graded against the 4 ppt limit, with the date.
2. Read your Consumer Confidence Report (CCR). Every utility publishes one yearly; it is the official local record.
3. On a private well? Test it. UCMR5 and CCRs do not cover wells – a certified lab test is the only way to know.

If your system is over the line, the next question is the filter that actually removes PFOA and PFOS: a certified NSF/ANSI 53 or reverse-osmosis unit, verified for your exact model in the Filter Certification Registry – not the marketing claim on the box.

How to comment on the EPA PFAS limits before July 20

Time-sensitive: the EPA accepted written public comments through July 20, 2026. If you are reading this after that date, the comment window has closed and the EPA is reviewing comments before finalizing the rules.

While the window is open, anyone can submit a written comment – you do not need to be an expert:

1. Go to regulations.gov.
2. Search the docket ID for the proposal you want to weigh in on: – EPA-HQ-OW-2025-1742 – keeping the PFOA/PFOS limits and extending compliance to 2031. – EPA-HQ-OW-2025-0654 – rescinding the PFHxS / PFNA / GenX limits and the Hazard Index.
3. Click Comment, write your view, and submit. Comments become part of the public record.

The July 7 virtual hearing’s sign-up to give spoken comment closed on July 1, so the written docket is the open path for most people. We are not going to tell you what to say – whether the limits should stay, the deadline should move, or the three others should keep their limits is your call. This is just where the decision is being made and how to be heard on it.

Related reading

• About 1 in 9 US water systems averaged PFOA/PFOS over the limit – our national data study, by state and by size.
• Which Filter Fits – find a certified PFAS filter for your exact system.

Sources

• EPA – proposed PFOA/PFOS compliance-extension rule
• EPA – proposed PFAS rescission rule
• regulations.gov
• EPA UCMR5