Industries / Food, farming and forestry / Food and beverage processing / Seafood processing
Food safety and tracing
This is the work of keeping processed seafood safe to eat and proving what it is and where it came from, from the landing to the shop.
7
Milestones achieved, from 1997 to 2023.
3
To watch for: announced, not yet confirmed outside.
2
Open deliverables. Nobody has shipped them.
The road
AchievedWatch forOpen
Achieved
1997
Hazard plans become mandatory
The US Food and Drug Administration required every seafood processor, including foreign plants shipping to the United States, to run a HACCP plan (hazard analysis and critical control points): find the steps where a hazard can enter, such as histamine building up in warm tuna, and check those steps every time, instead of only testing the finished product. The rule was published in December 1995 and took effect in December 1997.
Lessonthe plant now proves safety with its own records at each critical step.
Credit
US FDA; its magazine called it the first time HACCP was required for a US food commodity on an industry-wide basis
2010
Catch papers at the border
From 1 January 2010 the EU refused fish imports without a catch certificate validated by the fishing boat's flag state. Fish processed in another country also needs a processing statement linking the weight of fish that went into the plant to the weight of product that came out.
Lessonin 2023 the EU itself judged the paper scheme "not efficient" and moved it online (see Watch for).
Credit
Council of the European Union, Regulation (EC) No 1005/2008
2011
DNA barcodes identify fillets
Once fish is filleted, breaded or canned, its species cannot be seen. In 2003 Paul Hebert's group showed that one short stretch of a gene, called COI, can tell animal species apart like a barcode, and in 2011 FDA scientists published a laboratory method using it to identify fish "in a manner that is suitable for FDA regulatory use".
Lessonthe method was validated in one laboratory and for fish only, not shellfish.
Credit
Paul D. N. Hebert and colleagues, University of Guelph; FDA scientists led by Sara Handy
2013
One label in three wrong
The conservation group Oceana bought 1,215 fish samples in 21 US states between 2010 and 2012, tested their DNA, and found one-third mislabelled under FDA guidelines.
Lessonsingle surveys run high; a 2019 analysis of many studies put the typical mislabelling rate for a product at 8%, with a few products far worse.
Credit
Oceana
2015
Forced labour in peeling sheds
Associated Press reporters found migrants held in shrimp-peeling sheds in Samut Sakhon, Thailand, ripping the heads, shells and guts off shrimp bound for US and other supermarkets, and traced the shrimp to buyers. The reporting freed 2,000 people, by the Pulitzer board's account, and won the 2016 Pulitzer Prize for Public Service.
Lessonthe abuse sat with subcontracted peeling sheds, and one large exporter answered by hiring 1,200 peelers in-house.
Credit
Esther Htusan, Margie Mason, Robin McDowell and Martha Mendoza, Associated Press
2018
US import records required
The US Seafood Import Monitoring Program made importers of 13 species groups report harvest and landing data and keep records of each handover from harvest to US entry, starting with 11 groups in January 2018 and shrimp and abalone in December 2018.
Lessonby design it covers about half of US seafood imports.
Credit
NOAA Fisheries
2023
Forced labour in Chinese plants
The Outlaw Ocean Project reported that at least a thousand Uyghurs moved from Xinjiang under state labour programs had worked in seafood plants in Shandong since 2018, and later that fifteen plants had used more than a thousand North Korean workers. Buyers including Sysco cut ties with named processors.
Lessonthe many handoffs between boats, plants and exporters make the origin of seafood hard to track, and records that follow the fish did not show who cut it.
Credit
Ian Urbina and The Outlaw Ocean Project
Watch for
Digital catch certificates
The European Commission says CATCH, its online system for catch certificates and processing statements, mandatory for EU importers from 10 January 2026, offers "a fully digitised and paperless workflow". Exporting countries may still send paper, which importers then type in, and in October 2026 six non-EU national systems were linked to it.
Achieved whenthe main exporting and processing countries validate certificates and processing statements in CATCH or a linked system, and an evaluation shows digital cross-checks catching fraud that paper missed.
oceans-and-fisheries.ec.europa.euoceans-and-fisheries.ec.europa.eu
Claimed by
European Commission, December 2025
US traceability rule delayed
FDA's Food Traceability Rule will require firms handling listed foods, including histamine-forming finfish, smoked finfish, crustaceans and bivalves, to keep records at set steps, beginning with the first land-based receiving of fish from a boat. Its compliance date, first 20 January 2026, has moved to 20 July 2028 after Congress directed FDA not to enforce it before then.
Achieved whenFDA enforces the rule from July 2028 and seafood firms are shown producing the records within 24 hours during a real outbreak trace-back.
Claimed by
US FDA, page updated November 2026
Fast DNA chips for species
Researchers report a microfluidic chip, a card that moves tiny amounts of liquid through channels, that runs a quick DNA test and identified Atlantic cod, sablefish and toothfish in 141 retail products with 100% accuracy; 31.91% of the labels did not match the DNA.
Achieved whena second laboratory reproduces the results and a regulator or standards body validates the method for routine checks.
Claimed by
Xu Z., Chen J. and colleagues, Journal of Food Composition and Analysis, December 2025
Open
01
Labels checked against DNA
A public test set in which each row pairs a photo of a seafood label or menu entry with the DNA barcode result for the same product and a verdict against the legal market name, for testing tools that read labels and flag likely substitution.
The briefCredit · unclaimed02
Testing phone histamine readers
A public benchmark of phone-read histamine tests. Histamine builds up in tuna, mackerel and sardines kept warm and causes scombroid poisoning. Split samples of retail fish, fresh and canned, are each tested with a published phone-read sensor, a commercial rapid kit and the laboratory reference method.
The briefCredit · unclaimed
Open 01 · the brief
Labels checked against DNA
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public test set in which each row pairs a photo of a seafood label or menu entry with the DNA barcode result for the same product and a verdict against the legal market name, for testing tools that read labels and flag likely substitution.
Why it is still open
A 2022 review of seafood fraud calls for "a standardized and rigorous dataset through which food fraud can be scrutinized", and a 2019 analysis found "insufficient data" for most products, including all invertebrates. A 2021 study photographed every package it tested, but its data are "available on request", and searches found no public set that pairs label images with DNA results.
Built with what exists
A phone camera, COI barcoding by a commercial DNA sequencing service, the BOLD Systems identification engine, FDA's Seafood List of accepted market names, and an open vision-language model to read the label text.
First step
Ask the authors of three published barcoding surveys whether they kept package photos and would release them with their results.
Done looks like
500 products from at least 50 shops in two countries, each with photos, the DNA result and sequence, and a verdict, published under an open licence, with a baseline score for one label-reading tool.
Open 02 · the brief
Testing phone histamine readers
The first to publish is credited on this line, and the deliverable moves from Open to Achieved.
What to build
A public benchmark of phone-read histamine tests. Histamine builds up in tuna, mackerel and sardines kept warm and causes scombroid poisoning. Split samples of retail fish, fresh and canned, are each tested with a published phone-read sensor, a commercial rapid kit and the laboratory reference method.
Why it is still open
A July 2026 systematic review found most portable histamine platforms "still in the research or prototype phase" with "insufficient interlaboratory validation", and called for standardised data sets. Commercial rapid kits exist but need their own reader; Neogen's Reveal Q+ for Histamine is read on Neogen's Raptor reader.
Built with what exists
A phone-read histamine sensor from the published literature, such as the enzyme-based sensor described in Analytical and Bioanalytical Chemistry in 2023, Neogen's Reveal Q+ for Histamine, and liquid chromatography (HPLC) under EN ISO 19343 at an accredited partner laboratory.
First step
Buy tuna and mackerel, keep portions at room temperature for 0, 12 and 24 hours, freeze them, and send half of each to a laboratory while testing the other half with a phone-read sensor.
Done looks like
300 split samples covering fresh and canned tuna, mackerel and sardine, including samples left warm to build histamine; the phone reading's agreement with HPLC reported, as a pass or fail at the EU and US limits and as error in mg/kg; all paired readings published.