EPA PFAS Rule Changes 2026: What the Proposals Actually Do

On May 18, 2026, the EPA proposed two rules that would change the first federal drinking water standards for PFAS. The enforceable limits for PFOA and PFOS — 4.0 parts per trillion each — would stay in place, but water systems could request until April 2031 to meet them instead of April 2029, and the federal regulations for four other PFAS would be rescinded entirely.

Both of these are proposals, not final rules. As of July 2026 neither has taken effect, and the outcome is not settled. The EPA accepted written comments through July 20, 2026, and held a public hearing on July 7. What follows is what the proposals say, what they would change if finalized, and what a household can verify about its own water in the meantime.

KEY TAKEAWAYS

The 4.0 ppt limits for PFOA and PFOS are not being repealed. The extension proposal explicitly upholds them.
Eligible water systems could request two additional years — until April 2031 rather than April 2029 — to come into compliance.
Systems measuring PFOA or PFOS at or above 12 ppt would have to take short-term mitigation steps during that extension period.
A separate proposal would rescind the federal regulations for PFHxS, PFNA, HFPO-DA (GenX), and the Hazard Index mixture. The EPA’s stated basis is procedural, not toxicological.
Systems granted an extension must disclose it — through annual Public Notification and in the Consumer Confidence Report. That report is how a household finds out where its own system stands.

What the EPA Proposed on May 18, 2026

The two proposals are separate rulemakings with separate dockets. They are frequently discussed as one action, and they are not one action. The distinction matters, because they do different things to different compounds.

Proposal 1: Extending the PFOA and PFOS compliance deadline

The April 2024 final rule established a maximum contaminant level (MCL) — the enforceable legal ceiling — of 4.0 ppt each for PFOA and PFOS, and required water systems to comply by April 2029. The 2026 proposal (Docket EPA-HQ-OW-2025-1742) keeps those MCLs exactly where they are and changes the timeline instead.

Under the proposal, an eligible system could request two additional years, moving its compliance date to April 2031. The extension is not automatic and not universal — it is a request an eligible system makes. Two conditions attach to it. A system operating under an extension must notify its customers through annual Public Notification. And a system measuring PFOA or PFOS at or above 12 ppt — three times the MCL — must implement short-term mitigation actions to reduce exposure during the exemption period rather than simply waiting out the clock.

Monitoring and reporting timelines continue to follow the April 2024 rule. The proposal changes when treatment must be in place, not when testing happens. That separation is easy to lose in summary coverage, and it matters: a system operating under an extension would still be sampling on the original schedule, and those results would still surface in its annual reporting. The measurement obligation and the treatment obligation are being decoupled, not suspended together.

The 12 ppt trigger deserves a note of its own. It is set at three times the MCL, which means it does not function as a second, looser standard — it functions as a floor under the extension. A system between 4.0 and 12 ppt could use the full two years without additional obligations beyond notification. A system above 12 ppt could not simply wait.

Proposal 2: Rescinding the regulations for four other PFAS

The second proposal (Docket EPA-HQ-OW-2025-0654) would rescind the federal drinking water regulations for perfluorohexane sulfonic acid (PFHxS), perfluorononanoic acid (PFNA), hexafluoropropylene oxide dimer acid (HFPO-DA, commonly called GenX), and the Hazard Index — a combined standard covering mixtures of those three plus perfluorobutane sulfonic acid (PFBS).

The EPA’s stated rationale is procedural. The agency describes the proposal as necessary to “correct the unlawful procedure under which regulations for these PFAS were promulgated” under the Safe Drinking Water Act, and states that the proposal is “solely based on a need to correct this unlawful process.”

That framing is worth reading precisely. The EPA has not asserted that these four compounds became less harmful, and the proposal does not present new toxicological findings about them. It argues that the process used to regulate them was defective. Whether a rescission on procedural grounds is the right remedy is exactly what the comment docket exists to test — and it is not a question the underlying chemistry answers.

What Would Change, and What Would Not

Laid side by side, the two proposals affect different compounds in different ways:

Compound	April 2024 rule	If both 2026 proposals are finalized
PFOA	MCL 4.0 ppt, comply by April 2029	MCL 4.0 ppt, comply by April 2031 if extension granted
PFOS	MCL 4.0 ppt, comply by April 2029	MCL 4.0 ppt, comply by April 2031 if extension granted
PFHxS	Federally regulated	No federal limit
PFNA	Federally regulated	No federal limit
HFPO-DA (GenX)	Federally regulated	No federal limit
Hazard Index (PFHxS, PFNA, GenX, PFBS mixture)	Federally regulated	No federal limit

Two things follow from this table, and they pull in different directions.

The first is that the headline PFAS standard survives. The two compounds with the deepest toxicological literature and the widest detection footprint — PFOA and PFOS — keep their 4.0 ppt limits. Claims that the EPA is “repealing the PFAS rule” require context: the extension proposal upholds those MCLs in its own text.

The second is that the timeline moves and the coverage narrows. A system that would have needed treatment in place by April 2029 could lawfully operate until April 2031. For a household on a system currently above the MCL, that is potentially two additional years of exposure at the tap, with the 12 ppt mitigation trigger as the backstop for the worst cases. And for four compounds, no federal MCL would exist to be measured against at all.

Some states set their own PFAS drinking water standards, and a federal rescission would not automatically eliminate a state-level limit. Whether that applies to any given household depends entirely on the state — that is a question for a state’s own drinking water program, not one this article can answer generally. The general principle is that federal MCLs establish a floor rather than a ceiling: a state may regulate more strictly than the EPA, and several do. What a rescission would remove is the federal floor beneath those four compounds, which matters most in states that have not set their own.

It is also worth being precise about what a rescission would and would not do to information. Utilities that have already conducted PFAS monitoring under the Unregulated Contaminant Monitoring Rule have that data, and UCMR results are published regardless of whether a compound carries an enforceable limit. Removing an MCL removes the compliance obligation attached to a number. It does not delete the number, and it does not stop the sampling that produced it.

The Certification Gap the Rescission Exposes

This is where the rescission proposal has a practical consequence that is easy to miss, and it concerns filters rather than regulations.

When a filter is described as “certified for PFAS,” that phrase almost always means something narrower than it sounds. The relevant standard is NSF/ANSI 53, and the common certification covers PFOA and PFOS specifically (the legacy “P473” designation was folded into NSF/ANSI 53 in 2019, and some certifiers still print the older number). A product certified for PFOA/PFOS reduction has demonstrated performance against those two compounds under that protocol. It has not thereby demonstrated anything about PFHxS, PFNA, or GenX.

A smaller number of products carry a broader certification — listed as Total PFAS rather than PFOA/PFOS. The Pentair FreshPoint GRO-575B, for example, is separately IAPMO-certified to NSF/ANSI 53 for Total PFAS, and ProOne’s gravity units carry an IAPMO NSF/ANSI 53 listing for lead and Total PFAS. Those are different claims resting on different testing.

The distinction matters more if the rescission is finalized. The four compounds losing their federal limits are precisely the ones a PFOA/PFOS-only certification does not address. Under the 2024 rule, an MCL backstopped them regardless of what any household’s filter did. Without one, the coverage question moves entirely onto the product — and “certified for PFAS” is not a specific enough phrase to answer it.

Three points of hygiene apply to any certification claim, and they apply here with force:

Name the certifying body. NSF International, WQA Gold Seal, and IAPMO R&T are all legitimate ANSI-accredited certifiers to the NSF/ANSI standards. A WQA- or IAPMO-certified product will not appear in NSF’s own database, so absence from info.nsf.org is not evidence of absence of certification. “WQA-certified to NSF/ANSI 53” is accurate; calling that same product “NSF certified” is not.
“Tested to” is not certified. A manufacturer’s one-time lab result is not an ongoing certification with audits behind it. Marketing copy blurs these routinely.
NSF/ANSI 372 is a materials standard. It means lead-free construction. It is not evidence that a product removes anything.

For which products actually hold which certifications, and for what, see our guide to water filters for PFAS removal. For the underlying chemistry and health literature, see PFAS in drinking water.

How to Find Out Where Your Own System Stands

The proposals contain their own disclosure mechanism, and it is the most useful thing in them for an individual household.

A system operating under a compliance extension would be required to notify customers through annual Public Notification. Community water systems would additionally have to describe, in their Consumer Confidence Report, the reason for the exemption and the steps planned to reach compliance by April 2031. The CCR is the annual water quality report a utility is required to publish — most post it online, and most are required to deliver it to customers each year.

That means the question “is my utility taking the extension, and why” is answerable from a document the utility must produce. The CCR is also where a system’s most recent PFAS monitoring results appear. Reading it is the difference between knowing where a specific system stands and inferring it from national coverage.

Utility-level data has a structural limit worth naming. A CCR reports what was measured in the distribution system, not what comes out of one particular tap. For PFAS specifically that gap is narrower than it is for lead — PFAS enters at the source rather than leaching from household plumbing — but the way to resolve any individual question about a specific faucet is testing. Our guide to testing your water at home covers what home kits can and cannot establish, and when a certified lab is the only instrument that answers the question.

For readers in metros where PFAS is already the documented headline risk, the local reports have the specific numbers: Los Angeles, San Diego, Riverside, and Flint, among others.

What This Changes for a Household

The evidence supports a fairly narrow set of conclusions.

If these proposals are finalized as written, the PFOA and PFOS limits remain the law at 4.0 ppt, and the change is in how long a system has to get there. For a household on a compliant system, little changes. For a household on a system currently above the MCL, the practical effect is a longer interval before treatment is required — bounded by the 12 ppt mitigation trigger and disclosed through the CCR.

For PFHxS, PFNA, GenX, and the Hazard Index, finalization would remove the federal measuring stick. Nothing about that changes what is or is not in a given water supply. It changes what a utility is obligated to do about it.

Point-of-use filtration is the part of this a household controls directly, and it is worth considering for anyone whose CCR shows PFAS detections — with the caveat established above, that the specific certification scope, and the certifying body behind it, are what determine which compounds a product actually addresses.

Frequently Asked Questions

Is PFAS still regulated in drinking water in 2026?

Yes. The federal maximum contaminant levels of 4.0 ppt each for PFOA and PFOS remain in force, and the EPA’s 2026 extension proposal explicitly upholds them. A separate proposal would rescind the federal regulations for four other PFAS — PFHxS, PFNA, HFPO-DA (GenX), and the Hazard Index mixture. Both are proposals as of July 2026 and neither has taken effect. Some states also maintain their own PFAS standards independent of the federal rules.

Did the EPA repeal the PFAS drinking water rule?

No. That claim requires context. The extension proposal keeps the PFOA and PFOS MCLs at 4.0 ppt and changes only the compliance deadline, from April 2029 to a possible April 2031 for systems that request and receive an extension. The rescission proposal is narrower than “the PFAS rule” — it applies to four other PFAS and does not touch PFOA or PFOS.

What is the 2031 PFAS deadline?

It is a proposed compliance date, not a current one. Under the April 2024 rule, water systems must meet the 4.0 ppt PFOA and PFOS limits by April 2029. The 2026 proposal would let eligible systems request two additional years, moving their deadline to April 2031. Systems granted the extension must notify customers through annual Public Notification, and those measuring at or above 12 ppt must take short-term mitigation steps in the interim.

Why is the EPA rescinding limits for GenX and PFHxS?

The EPA’s stated rationale is procedural rather than toxicological. The agency describes the proposal as necessary to correct what it calls the “unlawful procedure” under which those regulations were issued under the Safe Drinking Water Act, and states the proposal is “solely based on a need to correct this unlawful process.” The EPA has not asserted that these compounds are less harmful than previously understood.

Does a PFAS-certified filter remove GenX and PFHxS?

Not necessarily. Most filters certified under NSF/ANSI 53 for PFAS are certified for PFOA and PFOS specifically, which is a narrower claim than “PFAS.” A smaller number carry a Total PFAS certification, which is broader. Because the four compounds in the rescission proposal are the ones a PFOA/PFOS certification does not cover, the specific certification scope — and which body issued it — determines what a product actually addresses.

Sources Cited

U.S. EPA, Proposed PFOA and PFOS Compliance Extension Rule — Docket EPA-HQ-OW-2025-1742
U.S. EPA, Proposed PFAS Rescission Rule — Docket EPA-HQ-OW-2025-0654
U.S. EPA, PFAS National Primary Drinking Water Regulation (final rule, April 2024)
NSF International certified product listings — info.nsf.org/certified/dwtu
IAPMO R&T product listings — pld.iapmo.org
WQA Gold Seal product listings — find.wqa.org

Filterd Water Guide