199 Food Recalls, 75 of Them the Serious Kind

US Tech Automations runs a daily clock against the FDA's own openFDA food enforcement feed and counts every distinct recall it finds — not a survey, not a sample, a running count of what the agency's own dataset shows. As of July 13, 2026, that clock had captured 199 distinct food recalls, deduplicated to one entry per recall number and counted using each recall's most recent observation.

Nothing in this report names a company, a brand, or a specific product. This is a counting report — how many recalls, how serious, and where the recalling firms are based — not a list of what to avoid buying today.

Key Takeaways

• 199 distinct food recalls captured as of July 13, 2026, spanning initiations from December 15, 2025 through June 9, 2026.

• 75 recalls (37.7%) are Class I — the FDA's most serious tier, meaning a reasonable probability of serious health consequences or death.

• 187 recalls (94.0%) are still open, meaning the FDA has not formally closed them — that is a paperwork state, not proof the food is still for sale.

• Idaho-based firms account for the most recalls of any state at 41, a 20.6% share — a reflection of where recalling companies are headquartered, not where the food was eaten.

• This feed covers FDA-regulated food only. Meat, poultry, and egg products are regulated separately by the USDA and are not counted here.

How Far Back This Edition Reaches

The 199 recalls in this edition were not all initiated in the same week. The earliest initiation date on file is December 15, 2025; the latest is June 9, 2026 — a span of nearly six months, because a recall can stay open in the FDA's system long after it was first initiated.

Of the 199 total, 152 recalls fall inside the trailing 90-day window ending on this edition's July 13, 2026 snapshot date. That is a separate slice of the data from the full 199-recall count above it, not the same number restated — a recall initiated in December 2025 and still open in July 2026 counts in the full total but not in the 90-day window, while a recall initiated and closed entirely within the last 90 days counts in both. Neither number is more "correct" than the other; they answer different questions about the same underlying dataset.

This pattern — one full count next to a separate, narrower recent-window count, each labeled for what it actually measures — is the same discipline behind our Search Console operating data, where a cumulative total and a trailing-window total are kept apart for the same reason.

What "Class I," "Class II," and "Class III" Actually Mean

These are not our categories — they are the FDA's own structured labels, applied by the agency at the time each recall is initiated, and this report leaves them exactly as assigned.

Class I recalls — the FDA's most serious tier — make up 37.7% of everything captured in this edition. No severity or health-risk judgment is inferred anywhere in this report; the class assigned to each recall is read directly from the FDA's own record and reported as-is.

Open, Terminated, or Completed: What Status Actually Tracks

Status is a separate question from severity — a Class III recall can sit open for months, and a Class I recall can already be terminated. The table below tracks only where each recall sits in the FDA's own closeout process.

94.0% of the recalls in this edition are still open in FDA paperwork terms. An open recall has not been formally closed by the agency — it does not mean the product is still on store shelves. In many cases the affected food was pulled from sale well before the FDA closes its file on the matter.

Recalls by the Recalling Company's Home State

This table ranks the 15 states with at least one recall in this edition, by the number of recalls tied to firms headquartered there.

Read the State Column Correctly, or Don't Read It at All

This is the single most important explanation in this report, so it gets its own section instead of a footnote. The state shown for each recall is where the recalling company is legally based — never where the food was sold, eaten, or where anyone got sick. Idaho tops this table at 41 recalls not because Idaho has a food-safety problem, but almost certainly because one or a small handful of large food producers or distributors headquartered in Idaho account for a disproportionate share of recalls in this window.

A single company's recalled products are very often shipped and sold nationwide, in every state — regardless of which state that company happens to be incorporated or headquartered in.

A state at the top of this ranking tells you where recalling companies are based. It does not tell you where recalled food was sold, where anyone got sick, or which state has the most contamination. Reading it any other way is the most common mistake made with recall data, and this report exists partly to head that mistake off.

Comparing two states on this table also means comparing however many firms happen to be headquartered in each — not comparing two states' food supplies, retail shelves, or safety records against one another. A state with zero recalls in this edition is not a state with a cleaner food supply than the fifteen listed here; it is simply a state with no recalling firm headquartered in it during this window, which is a fact about company addresses, not about food safety anywhere.

Method in Plain English

One recall is one FDA recall_number — a single administrative action that can cover one product or, just as easily, dozens of related products sold under different labels, sizes, or dates. These are counts of recalls, not counts of individual items pulled from shelves, so a state or class total here should never be read as a count of affected products. Each recall is counted once, using its most recent observation, so a recall whose status updates daily in the FDA's feed is not double-counted just because the agency re-published it.

Every count in this report is a verbatim count of distinct FDA recall_numbers our own clock actually captured; nothing is estimated, modeled, or inferred. No cause, severity judgment, or health-risk assessment is added beyond the FDA's own classification. This edition covers FDA-regulated food only — meat, poultry, and egg products fall under USDA/FSIS jurisdiction and are tracked in a separate federal system, not this one.

This report re-shapes one already-sealed capture of the FDA-enforcement clock rather than re-collecting anything itself; nothing was fetched fresh to build this page. Our formation-index report applies the same re-shape-a-sealed-capture method to a different first-party dataset, if that pattern is useful context.

Put the Recall Data to Work

A food-safety compliance lead or a regulatory-affairs manager tracking exposure across a supply chain cannot get a same-format, dated count like this from the FDA's own portal without building a scraper and a classification pipeline first. US Tech Automations runs the daily clock, applies the FDA's own class and status labels without reinterpreting them, and reshapes the result into one dated, sealed report every month instead of someone re-running a manual pull each time the question comes up. Our permits-by-metro roundup applies the same daily-clock discipline to residential construction, if a second first-party dataset built the same way is useful.

Frequently Asked Questions

Q: How many food recalls are open right now?
A: As of this edition's July 13, 2026 snapshot, 187 of the 199 distinct recalls captured are still open in FDA terms — meaning the agency has not formally closed them. Open does not mean the food is still on sale; it is a paperwork state.

Q: How many of the recalls are the serious kind?
A: 75 recalls, a 37.7% share, are Class I — the FDA's own most serious tier, defined as a reasonable probability of serious health consequences or death. These are the FDA's own classifications, not a severity judgment made by US Tech Automations.

Q: Does Idaho topping the state ranking mean Idaho has a food-safety problem?
A: No. The state column shows where the recalling company is legally based, not where the food was sold or eaten. Idaho's 41 recalls almost certainly reflect a small number of firms headquartered there whose products are shipped and sold nationwide, not a problem specific to food produced or sold in Idaho.

Q: Does this cover all food recalls in the United States?
A: No. This edition covers only the FDA's food enforcement feed. Meat, poultry, and egg products are regulated by the USDA under a separate system (FSIS) and are not included in this count.

Q: Can one recall count as more than one product?
A: Yes, routinely. One recall number can cover dozens of separate products across different labels, sizes, or production dates. This report counts distinct recalls, not the number of individual products or items affected.

Method and Provenance

Every figure in this report is read directly from our own sealed FDA-enforcement clock, re-shaped from an already-sealed edition dated July 13, 2026. This report covers the U.S. FDA's openFDA food enforcement feed only, deduplicated to 199 distinct recall events — not USDA/FSIS meat, poultry, and egg recalls, and not drug or device recalls. Every count is a verbatim count of distinct FDA recall_numbers the clock actually captured; nothing is estimated, modeled, or inferred beyond the FDA's own class and status labels.

Source: US Tech Automations Research — FDA Food Recalls series, sealed 2026-07 edition, from our own FDA-enforcement clock (openFDA food enforcement feed).