Accurate, verifiable data is the foundation of sustainable fisheries management, yet much of it is still gathered and checked by hand. In many fisheries, catches are logged manually, compliance checks are labour intensive, and hours of onboard footage still depend on painstaking manual review.

Many governments are introducing increased fisheries regulation. In Scotland, remote electronic monitoring (REM) has been mandatory for scallop dredgers since 17 June 2024. And since 7 March 2026 it has also applied to pelagic boats of 12 metres+ with chilled or freezer storage. In England, Defra is introducing REM across five priority fisheries. In the EU, by 2028 the revised Fisheries Control Regulation will require REM, including CCTV on vessels of 18 metres+ that pose a high risk of non-compliance. New Zealand has fitted cameras across its inshore fleet in what its seafood industry describes as the largest commercial fisheries camera programme in the world.

Cameras and sensors are only half the story. As a UK Government blog wrote after England’s first REM trial: “fitting cameras is the easy bit”. The real challenge lies in turning large volumes of video, sensor and positional data into reliable and accurate decision making. A task that manual review cannot scale to meet. This is where AI and computer vision can deliver a step change, bringing benefits for a range of stakeholders:

• Healthier fish stocks: catch data underpins better stock assessments and fairer quota decisions, helping to safeguard catches for the long term.
• Healthier marine ecosystems: greater visibility of gear and fishing activity helps to reduce bycatch and lost “ghost” gear.
• More time fishing, less time on paperwork: automated records mean crews have substantially less paperwork to do, reducing administrative burden.
• Trusted, traceable seafood: verified catch data from monitoring systems gives buyers, retailers and consumers increased confidence in sustainable sourcing.