BTC
ETH
SOL
XRP
BNB
TON
DOT
LTC
POL
HBAR
NEAR
ARB
UNIPolice test Graylark AI that turns interior photos into home addresses
Raven, built by Graylark Technologies, matches room details against real estate and rental listings to infer an address without GPS or metadata. The LAPD and Miami-Dade Sheriff's Office are among agencies testing it.
Why it matters: The tool shows how ordinary interior images can be turned into precise locations, raising privacy and legal questions for the rental and property platforms whose listings feed it.
Key points
- Raven matches fixtures, cabinetry and layout geometry to photos on sites including Zillow, Airbnb and Vrbo.
- Demos linked a laundry room to a Charlotte Airbnb and a bathroom to a Montecito Vrbo rental, per reporting cited by 404 Media.
- Airbnb says it has no relationship with Graylark and strips location metadata; Zillow warns unauthorised scraping breaches its terms.
Graylark Technologies built the system, which matches details inside a room against real estate and short-term rental listings online. It needs no GPS, photo metadata or outdoor landmark to work.
Raven compares fixtures, cabinetry, moulding, appliance models and the geometry of a layout with large collections of labelled property photos. A bedroom photo reportedly led back to a San Francisco address.
The scraping behind the tool has drawn pushback. Zillow says unauthorised extraction breaches its terms and that it can take legal action. Graylark restricts Raven to law enforcement and vetted large customers, after an earlier version known as GeoSpy was more widely available and reporting described misuse.
The tool has limits: it depends on publicly available labelled images, so it can fail where a home has never been listed or has changed substantially. Internal emails cited in reporting suggest the Miami-Dade office used a version tailored to its county rather than a universal tool.
Live updates on NewsFeeds:AIDesk editor: Keshav Aggarwal (accountable for this page; this summary was not edited by hand)
How this was produced: an AI model wrote this summary of Bitcoin.com News’s report, which we credit and link above. Read our Editorial and AI policy. Spotted a mistake? See Corrections.
Not investment advice. Read the risk disclaimer.