bilateral
TopicsAbout← Feed
BySRSam Reyes·CMCal Morrow·EQEliza Quinn·DPDana Park
BREAKINGMay 3, 2026

“A Huge Setback”: New EPA Directive Could Weaken Hundreds of Chemical Regulations

EPA Deputy Administrator David Fotouhi issued an internal memo sharply criticizing the agency's Integrated Risk Information System (IRIS) program and directing EPA offices to review the chemical assessments they have relied upon. The memo effectively signals that the library of more than 500 IRIS chemical assessments cannot be trusted, opening the door to weakening hundreds of chemical regulations, permits, and enforcement actions at both the state and federal level. Before his appointment, Fotouhi worked as a lawyer representing companies accused of causing toxic pollution.

XLinkedInFacebookThreadsWhatsAppReddit

Should the EPA prioritize removing outdated chemical rules to speed up business, or does rolling back protections risk exposing people to substances we now know are dangerous? The agency just picked a side.

Not familiar with this story? Get context →
Memo creates litigation crowbar for polluters
Liberal
Fotouhi's memo doesn't just criticize IRIS — it hands every company facing a chemical enforcement action a ready-made legal argument that the science underneath their case was always questionable, because the EPA's own deputy administrator said so. That mechanism is not subtle and not incidental. Any regulated party can now point to this memo and litigate a permit, a pollution limit, or an enforcement action into paralysis — and regulatory paralysis benefits exactly one side in a chemical dispute, and it is not the community downstream.
Conservative
You're describing a lock the locksmith can pick — but that's an argument for changing the lock, not for declaring it unpickable forever. If IRIS assessments can be unraveled in court by a single internal memo, that tells you something important: the methodological foundations were never as airtight as the agency claimed. The answer to legally fragile science is more rigorous science, not institutional immunity from challenge.
Liberal
More rigorous science requires a replacement standard, a funded successor program, transition rules — none of which the memo proposes. You've just restated the problem: legally fragile science with nothing behind it is not reform, it's a void, and voids benefit the party with lawyers.
Conservative
A void is preferable to a false floor — regulations that appear solid but crumble under scrutiny impose real costs while offering only the illusion of protection. The pressure to fill that void is exactly what should force a genuine scientific reckoning.
No replacement standard offered
Liberal
If this were honest regulatory reform, you name the replacement standard first. You convene independent scientists, fund a successor program, propose transition rules. What the memo does instead is declare 500-plus assessments invalid and offer nothing — no alternative methodology, no replacement clearinghouse, no timeline. That sequencing isn't an oversight. The void is the point.
Conservative
You're treating the absence of a replacement as proof of bad faith, but that's backwards. IRIS didn't arrive with a fully formed successor to what it replaced — it was built incrementally because the need for better science drove the institution. Declaring the current system inadequate is the necessary first step; you can't architect a replacement while the existing framework is treated as untouchable.
Liberal
IRIS was built into a space that had no enforceable standards — Fotouhi is clearing a space that currently does. Those are not equivalent situations. Tearing down an existing standard while communities drink water from pipes regulated under it is not a prerequisite for reform; it's the harm itself.
Conservative
If the standard is scientifically indefensible, enforcing it isn't protection — it's the appearance of protection. The uncomfortable question isn't whether a replacement exists yet; it's whether some of these limits were ever defensible enough to impose compliance costs in the first place.
GAO criticism predates current complaints
Liberal
The pattern around this memo makes intent impossible to dismiss. PFAS rules weakened in November 2025, the 2024 chemical risk management rule rescinded, ethylene oxide deregulated, public access to chemical hazard data restricted — and now all 500 IRIS assessments cast into doubt simultaneously. Project 2025 called explicitly for IRIS's elimination using language that had already appeared in industry-backed litigation for years. This is a pre-planned sequence, not regulatory housekeeping.
Conservative
The GAO flagged IRIS's methodology, transparency failures, and decade-long update delays back in 2011 — that's not industry talking points, that's the government's own watchdog. If the pattern you're describing is 'finally acting on longstanding legitimate criticisms,' that looks identical from the outside to what you're calling a conspiracy. The chronology doesn't establish motive.
Liberal
The GAO criticized process delays and transparency — it did not recommend invalidating 500 assessments at once, and it certainly didn't recommend doing so under a deputy administrator who spent his career litigating against the science he's now reviewing. Context isn't conspiracy; it's evidence.
Conservative
Conflict-of-interest optics are genuinely bad — I said so. But 'this person had industry clients' doesn't make the underlying criticism of IRIS wrong, any more than a defense attorney's client list makes the charges they contest automatically valid.
Who bears regulatory costs and benefits
Liberal
When you hear 'billions in economic costs' from IRIS-based rules, the honest question is: costs to whom, compared to what? Industry-produced cost estimates systematically omit the economic value of avoided cancers, reduced hospitalizations, and recovered productivity. The communities absorbing the risk when those limits are lifted are disproportionately low-income, disproportionately Black and Latino — living near the facilities that lobbied for exactly this outcome.
Conservative
You're right that industry cost estimates omit health benefits — but EPA benefit estimates have their own thumb on the scale, assigning dollar values to avoided cancers using statistical life valuations that most affected communities never consented to and couldn't interrogate. Both sides of this ledger involve contested assumptions. The difference is that compliance costs fall on identifiable firms right now, while projected health benefits are probabilistic claims about populations over decades.
Liberal
Probabilistic claims about cancer over decades are exactly what toxicological science is for — the alternative you're implying is waiting for bodies to count. The arsenic standard was weakened in 2001 under the same 'contested assumptions' argument and reinstated because the harm was real. We've run this experiment.
Conservative
The arsenic standard was reinstated — through a democratic process that weighed the evidence and changed the rule. That's the system working, not proof that the original standard was beyond challenge.
Institutional authority versus scientific legitimacy
Liberal
IRIS has served for four decades as the recognized standard for chemical toxicity used by regulators from Sacramento to Stockholm. Casting doubt on all of it at once doesn't make science more rigorous — it makes the EPA's own scientific infrastructure untrustworthy to every regulator, court, and trading partner that built on it. The damage to that institutional credibility is not recoverable in a news cycle.
Conservative
An institution's age and adoption breadth are not arguments for its accuracy. IRIS assessments have often gone a decade without updates, operating outside the external peer review norms that govern the rest of science. Describing that as 'settled science' confuses how long something has been on the shelf with whether it still holds. International adoption of flawed baselines doesn't make those baselines correct — it makes the error bigger.
Liberal
External peer review of individual contested assessments is a legitimate ask. Declaring all 500 simultaneously questionable — through an internal memo, not a scientific process — is not peer review. You're describing a standard of rigor the memo itself doesn't meet.
Conservative
The memo is an administrative directive to begin a review process, not a scientific finding. Treating the initiation of scrutiny as the harm is precisely the institutional reflex that made IRIS unaccountable for forty years.
Conservative's hardest question
The administration has articulated no credible, funded alternative scientific framework to replace IRIS assessments if they are broadly invalidated — meaning the practical consequence of this memo is not better science, but a regulatory vacuum where polluters can litigate standards into paralysis while communities bear the harm.
Liberal's hardest question
The strongest challenge to this argument is that some IRIS assessments do rely on contested dose-response extrapolations and conservative assumptions that independent toxicologists have questioned on scientific grounds — meaning a blanket defense of every IRIS assessment as settled, unimpeachable science is harder to sustain than the argument that the process of invalidating them wholesale, without a replacement standard, is corrupt. The distinction matters: the case against Fotouhi's memo is strongest on process and conflict-of-interest grounds, and somewhat more vulnerable on the claim that every IRIS number is beyond legitimate scientific scrutiny.
The Verdict
Both sides agree
Some IRIS assessments rest on dose-response modeling assumptions and epidemiological extrapolations that are genuinely contestable on scientific grounds, not merely industry complaints — both sides concede this even if they disagree sharply about what to do with that fact.
The real conflict
FACTUAL: Whether the GAO's 2011 process criticisms of IRIS (speed, transparency, delays) support the scope and simultaneity of Fotouhi's 2026 memo invalidating 500+ assessments at once — conservatives cite this as precedent for systemic review, liberals argue it justified targeted methodological reform, not wholesale doubt-casting.
What nobody has answered
If IRIS assessments have been scientifically adequate enough to serve as the basis for international regulatory standards for four decades, what specific evidence would the EPA need to present to demonstrate that invalidating them wholesale without a replacement standard is actually about improving science rather than achieving what industry litigation has pursued for years — and how would anyone distinguish between the two?
Sources

More debates