UPS deploys AI to fight porch pirates
UPS just expanded DeliveryDefense, an AI tool designed to predict which doorstep deliveries are likely to get stolen. The model crunches billions of delivery records to score addresses from 1 to 1,000, allowing merchants to reroute sketchy drops to pickup lockers before theft occurs. With holiday theft peaking, UPS is betting algorithmic risk scoring beats doorbell cameras.

Porch pirates stole 260 million packages across the U.S. last year. Even worse: smart doorbell cameras barely scare them away, with research showing only 5% of thieves even bother hiding their faces.
Now, UPS wants to stop the theft before the box even leaves the warehouse.
Rating doorsteps on a 1,000-point scale
UPS Capital rolled out DeliveryDefense, a machine learning tool that assigns delivery addresses a "Delivery Confidence Score" from 1 to 1,000.
A score of 1,000 means the package is almost certain to arrive safely. A score near 100 flags an address as a high-risk theft zone.
The algorithm crunches billions of data points collected from two years of UPS delivery history across the country. UPS says it analyzes physical delivery characteristics rather than individual customer profiles—looking at whether a building has a locked mailroom and chain of custody, or if packages are left out in the open.
Shifting the risk before shipping
The tool is designed for merchants, not end-consumers. Enterprise API access starts at $3,000 a month, which was recently tested with Costco Wholesale in Colorado. A lighter web dashboard for small businesses launched ahead of the holiday shipping rush at $99 per month.
When a customer’s address flags a low score, merchants can take action before shipping. They can redirect the parcel to an access point like a local CVS, require a physical signature, or buy additional shipping insurance.
While Amazon pushes garage deliveries and DHL sticks to old-school signatures, UPS is the first carrier trying to predict doorstep crime through raw data.
Why it matters
We think DeliveryDefense is a clever pivot from hardware deterrence to predictive risk management. Cameras capture theft after it happens; predictive scoring stops the package from being abandoned on an open porch in the first place.
For online shoppers, especially apartment dwellers in high-density areas, there is an uncomfortable downside. Your address might get slapped with a low confidence score through no fault of your own, forcing you to hike to a retail pickup locker instead of enjoying doorstep delivery.
For merchants selling high-ticket electronics or apparel, the $99 monthly fee is an easy hedge against stolen inventory.
The real test will be algorithmic fairness. If DeliveryDefense consistently downrates entire zip codes or lower-income neighborhoods, UPS and its retail partners will face immediate backlash over digital redlining.
Sources (1)
- cnbc.com(opens in a new tab)Originalcnbc.com