中文Every label converter knows the moment. The press has run clean, Helios has logged the defects, and now a human has to sit and review the virtual roll — frame by frame, defect by defect — deciding what is salable and what gets pulled at the rewinder. It is the most important step in quality, and the most tedious job on the floor. AVT just built an AI to take it off your team’s plate.
AVT launched AutoEdit AI, making its AI-driven post-inspection defect-classification module available to the wider label market on a subscription basis for AVT Helios 100% print inspection systems. The module debuted at Labelexpo Europe 2025 and has been running on real converter production floors since. It targets the bottleneck that appears after detection: the manual review and classification of every logged defect.
Only the hard calls reach a human
Here is how it works, and why it is clever. AutoEdit AI runs after Helios inspection and before AVT Workflow Link automatic waste removal. It classifies every detected defect into one of three buckets: salable, non-salable, or requiring final review. Only that third category — the genuine edge cases — surfaces to the operator. The pre-sorted view lands inside PrintFlow Manager, the interface operators already use, so there is no new screen and no new station to learn.
The model was trained centrally by AVT across production data from many converters, applications, substrates and end-market quality standards. Every classification and every operator override is logged in PrintFlow, creating an audit trail — exactly what pharmaceutical, food, healthcare and personal-care converters need when a customer asks ‘how do you know this was fine to ship?’ The module ships with both software and dedicated processing hardware, installs alongside existing Helios systems without disruption, and receives periodic updates carrying accumulated industry learning.
The pressure after detection
Guy Yogev, VP of marketing and product at AVT, named the real problem: ‘Detection has been a settled question in our industry for years. What has changed is the pressure on everything that happens after it.’ As short runs, versioning and personalisation multiply, that manual review of the virtual roll has become ‘one of the real constraints on throughput, manpower and consistency between operators and between shifts.’ AutoEdit AI is designed to lift that classification burden directly.
Roy Porat, CEO of AVT, framed it as the company’s next chapter: ‘deeper into the intelligence layer of print inspection and closer to the day-to-day realities of the converter’s shop floor.’ One early adopter described the impact in plain language: ‘AutoEdit AI removed a step our team used to do manually on every roll. We’re catching more, editing less, and our operators are freed up to focus on the jobs that really need their expertise.’
Why this matters beyond the spec sheet
Quality is where trust is won or lost in label printing. But trust built on a human reviewing thousands of frames per shift is fragile — people tire, shifts differ, standards drift. An AI that handles the routine and escalates only the genuine judgement calls makes the whole line more consistent, and frees your most expensive people to do the work that actually needs a brain.
The emotional win is dignity, not just efficiency. Nobody got into print to stare at defect thumbnails for eight hours. Give that job to the machine, and you give your operators back their judgement. That is the kind of automation worth cheering.
Source: Labels & Labeling, 8 September 2026.
