Context
Winnow Vision combines a camera, smart scale, tablet, and recipe-cost attribution so kitchens can see what is being wasted and where the cost is coming from. The AI measures; the humans act.
The deck's anchor case is Hilton Dubai · Al Habtoor City: the property reports roughly 70% food-waste reduction and $65K annual savings from a production-kitchen programme. Mandarin Oriental, Emaar, and Hilton Green Ramadan sit downstream as adjacent Winnow patterns.
What the AI actually does
At Hilton Dubai, Winnow Vision captures production-kitchen waste through the camera + scale + tablet workflow. The model classifies discarded food, the scale records weight, and the recipe-cost layer turns that waste into a daily cost signal.
The chef and F&B director then change menus, portions, prep batches, and purchasing. Hilton pairs the tech with behaviour-change coaching sessions, and during Green Ramadan 2025, partnered with UNEP West Asia for the "Recipe of Change" campaign. The dashboard is the dashboard, not the fix; the operational change is human.
Measurable outcomes
- ~70% food-waste reduction across production kitchens at Hilton Dubai Al Habtoor City
- $65K annual savings at the property
- Daily item-level cost signal that previously did not exist
- Mandarin Oriental 4-property pilot: 36% reduction · $207K annualised group savings · 66 tonnes avoided
- Hilton Green Ramadan 2025: 26% plate-waste reduction across 45 hotels in 14 countries
What to copy
Computer vision on a single, high-cost waste stream: measure first, then act. The ROI story is concrete (food cost · CO₂ · hours · bin weight).
The operational discipline matters as much as the tech: weekly behaviour-change coaching, chef-owned response loop, and "you can't reduce what you don't measure" as the implicit thesis.
What doesn't transfer
The ROI depends on high-volume F&B operations. A 30-cover restaurant will not produce Mandarin-scale numbers, the absolute waste pool is too small.
Concentration risk is real: a single-vendor story is an operating pattern, not a vendor endorsement. Mandarin's pilot ran 6 months before group-wide rollout, operators should expect a measurement-then-behaviour-change runway, not instant savings.
Open questions before buying
- Camera placement and bin geometry, can your kitchen accept the hardware without disrupting service flow?
- Recipe-cost database, is yours up to date and machine-readable? The dashboard is only as good as the costs it pulls from.
- Kitchen culture: do your chefs respond to a daily numbers-on-the-wall feedback loop, or will the dashboard sit unread?
- Payback timeline: published numbers reflect mature deployments; the first three months are typically a baselining phase, not a savings phase.
The vendor, Winnow
Winnow ships AI vision systems specifically for back-of-house kitchen waste, it does not pretend to be a general-purpose food-tech platform.
The company reports a vendor-aggregate $100M+/year savings across all operators in 2026, 28,000 tonnes of food waste avoided, 122,000 tonnes of CO₂e. Treat as scale signal, not a single-property number. Concentration risk: in 2026, hospitality has effectively one named AI-vision food-waste vendor.