- 1 min read
- Third edition
AI robots handle Zalando fashion returns in Germany, Poland
CEVA, Zalando and Sereact run AI robots on fashion returns at two live European sites, but no productivity result is yet published.

Market Context & Direct Impact
CEVA Logistics, Zalando and Sereact are now running AI-powered robots on fashion returns in Germany and Poland. Cortex, Sereact's Physical AI platform, drives dual-arm robots at sites in Greven, Germany, and Świebodzin, Poland, according to Container News. The launch is the first operational phase of a wider partnership to expand automated returns processing across Europe. Partners deploy the machines as a service rather than having buyers own the equipment outright.
The operating result is still undisclosed, so the metric that matters remains unproven. CEVA's Chris Walton called returns handling one of the toughest challenges in contract logistics, citing unpredictable volumes and a disproportionately high cost per unit. The companies have not disclosed how many robots are deployed or what productivity gains they have achieved. The model's appeal is that it can grasp items it has never seen, without per-SKU training, which matters when returned clothing arrives in irregular condition. Separately, Bloomberg reported that Sereact raised $110 million in a Series B round led by Headline. That funding describes the vendor's growth, not this deployment's performance.
Core Executive Takeaways
- No figure yetThe launch discloses no robot count or productivity gain, so any business case needs its own pilot baseline.
- Unseen-item handlingCortex is designed to grasp products it was never trained on, which suits mixed fashion returns.
- Service-based contractsRobotics-as-a-Service lets buyers access automation without owning the equipment, changing capex into a running cost.
Strategic Playbook
- Set the metric first
Before signing, define cost per returned unit and handling time as pass-or-fail measures, and require the vendor to share its manual baseline.
- Scope a returns pilot
Run one irregular-returns stream on a fixed-term service contract, and compare results against manual handling at the same site.
Filed 21:26 GMT, 14 October 2026, for the third edition of 14 Oct.
As agents only.
Sources
- CEVA Logistics and Zalando deploy AI-powered robots for returns handlingContainer News, published 9 October 2026
- AI Startup Sereact Raises $110 Million for Robots That Predict ConsequencesBloomberg, published 27 April 2026
Claims ledger: 11 facts and where each comes from
- CEVA Logistics, Zalando and Sereact launched AI-powered robotic systems to automate fashion returns at logistics facilities in Germany and Poland.1
- The systems are operating at sites in Greven, Germany, and Świebodzin, Poland.1
- The robots use Cortex, Sereact's Physical AI platform, to handle different types of returned products.1
- The technology can identify and grasp previously unseen items and does not require separate training for each product or fixed SKU profiles.1
- The robots are deployed through a Robotics-as-a-Service model, under which partners provide robotic capability as a service rather than relying solely on equipment ownership.1
- CEVA's Chris Walton said returns handling is one of the toughest challenges, given unpredictable volumes, manual quality checks, seasonality and a disproportionately high cost per unit.1
- The deployment marks the first operational phase of a wider partnership to expand automated returns processing across Europe.1
- Zalando joined Sereact's Series B funding round as a strategic investor in July 2026.1
- The companies have not disclosed the number of robots deployed or the specific productivity improvements achieved so far.1
- Sereact raised $110 million in fresh funding to develop its AI model that makes robots smarter and more adaptable to different tasks.2
- The Series B funding round was led by venture capital firm Headline, with other new investors including Bullhound Capital, Felix Capital and Daphni.2
Every figure above is tied to a cited source that was checked on the day of filing. Found an error? Tell the desk.


