MIT is spending more than $3 million to install over 500 AI-capable video cameras across academic buildings, residence halls, and outdoor spaces along Memorial Drive, with installation that began in November 2025 and is expected to continue until September 2026.
MIT's $3 million procurement and installation timeline
The university has committed "over $3 million" for a deployment of more than 500 cameras, according to reporting. Installation, wiring, and the supporting infrastructure began in November 2025 and will likely continue through September 2026. The stated deployment locations include academic buildings, residence halls, and outdoor areas along Memorial Drive, signaling a broad campus footprint for the new system.
Hanwha Wisenet AI cameras: resolution, recognition, and movement
Nearly all of the new cameras are part of Hanwha’s Wisenet AI line. Those cameras are marketed for their ability to identify and classify multiple objects using deep learning algorithms, supporting resolutions that range from 2MP to 4K. The product descriptions cited indicate real-time recognition of faces, license plates, vehicles, and other objects, and most units will support pan, tilt, rotate, and zoom motion.
Classification and detection features in real time
Technical specifications reported for the cameras show they are capable of collecting real-time face and object classification data and detecting a set of behaviors and conditions, including motion, loitering, crowds, face masks, and camera tampering. The cameras can automatically classify individuals on the basis of clothing color, gender, and age. The reporting includes a specific range: classification of individuals is possible up to a distance of 35 feet (11 meters) from the camera.
Continuous monitoring with Ai‑RGUS
Nearly all cameras will be monitored continually with Ai‑RGUS, described as an AI camera software. That continuous monitoring capability is noted alongside the cameras’ pan/tilt/zoom functions and the devices’ real-time recognition features, indicating the system is designed for active, ongoing automated observation rather than intermittent or solely archive-based review.
Data retention: "retained up to 30 days," exceptions allowed
MIT spokesperson Kimberly Allen is quoted as saying any collected data is "retained up to 30 days," unless an exception is granted. The reporting does not enumerate what circumstances would trigger an exception, but it records MIT’s stated baseline retention period of 30 days for collected camera data.
What this means for students, campus security, and Hanwha/procurement
- Students and residents: Individuals on campus will be subject to automated classification by clothing color, gender, and age, and to detection of motion, loitering, crowds, and face masks within the cameras’ effective range (up to 35 feet / 11 meters).
- Campus security and facilities: Security operations will receive feeds from cameras designed for real-time recognition of faces, license plates, vehicles, and other objects, and those feeds will be monitored continually with Ai‑RGUS; the installed hardware also includes pan/tilt/zoom capability and high-resolution (2MP–4K) imaging.
- Hanwha and procurement teams: The deployed units are Hanwha Wisenet AI models promoted for deep-learning classification features; procurement reflects purchase and infrastructure work that began in November 2025 and is slated to continue through September 2026.
The facts in the record are concrete: hundreds of AI-capable cameras, marketed to perform live face and object recognition at resolutions up to 4K, are being installed campus-wide; installation and wiring began in November 2025 and are expected to continue until September 2026; monitoring will be continual via Ai‑RGUS; and MIT has stated that collected data is "retained up to 30 days," with unspecified exceptions. Together, those specifics define the campus surveillance posture at least through the stated installation window and frame the questions that remain about how exceptions to retention will be applied and how continual AI monitoring will be operationalized.
Original reporting: https://www.schneier.com/blog/archives/2026/07/mit-to-become-hotbed-of-ai-video-surveillance.html
