A look at surveillance data fusion, which happens when thousands of disparate devices and tools become one all-seeing surveillance platform
Arthur Holland Michel / Wired : Tweets: @torproject , @wired , @gadgetlab , @wired , and @hypervisible Tweets: @torproject : “We're rarely not in the crosshairs of some spying device... The fact that intelligence can be difficult and tedious to correlate was perhaps the last natural rampart standing between us & total surveillance... Fusion technology eviscerates those spaces.” https://www.wired.com/... @wired : With fusion technology, our digital footprints, once scattered, become a single uninterrupted life history. https://www.wired.com/... @gadgetlab : Security cameras. License plate readers. Smartphone trackers. Drones. We're being watched 24/7. What happens when all those data streams fuse into one? https://www.wired.com/... @wired : The idea of an ever-expanding, all-knowing surveillance platform used to be a technologist's fantasy. The emergence of multi-intelligence fusion might make it a reality https://www.wired.com/... @hypervisible : Harrowing long read. You probably don't want to know what a “correlation engine” is in a LE context, but you need to know. https://www.wired.com/...
Context & Ripple Effects
This Wired piece by Arthur Holland Michel names the layer that the prior coverage was building toward: after years of visual surveillance hardware spreading and then AI-enabled camera networks piping private feeds to police headquarters, fusion is the software step that correlates it all — turning scattered device outputs into one searchable record.
The later coverage confirms the trajectory Michel described: Fusus now links thousands of US locations' cameras into a single hub with AI analysis, exactly the consolidation-from-disparate-sensors pattern this article flagged, while Schneier's argument about AI enabling mass spying explains why correlation stopped being the bottleneck.
First-order effects
- Police departments running fusion platforms like Fusus gain cross-source search over public and private cameras in one interface, collapsing what was previously separate investigations per feed into single queries.
- Private camera owners — businesses and homes already volunteering feeds to police hubs — become involuntary contributors to a unified surveillance record they do not control.
Second-order effects
- Vendors compete on integration breadth rather than sensor quality, pushing Fusus-style platforms to sign more municipalities and private partners to make their index the one police query first.
- Civil-liberties pressure shifts from individual devices (a camera, an ALPR) to the fusion layer itself, since challenging one sensor no longer limits what the combined record reveals.
Third-order effects
- If correlation is the last rampart, as the Tor Project framing in the piece argues, regulation aimed at collection becomes insufficient and governance has to target aggregation and search — a structural shift from device-level privacy law to platform-level rules.
- The market consolidates around a few fusion vendors holding the cross-agency index, making procurement decisions by individual towns consequential for regional surveillance capability.
The trend: Surveillance is moving from scattered sensors to fused platforms where AI makes correlation cheap, shifting both vendor competition and privacy fights from collection to aggregation.