Logitech Rally AI Camera Pro assessed: a $2,999 camera that is also a room sensor

Logitech introduced the Rally AI Camera and Rally AI Camera Pro in January 2026 and used LEAP in Riyadh to put the Pro in front of the regional market. This is an assessment of the published specifications. TechScoop has not installed either camera, has not sat in a meeting through one, and makes no claim about image quality, framing accuracy or reliability in a real room. The specification is the strongest part of the published material. The Pro is a dual-camera system with 15x hybrid zoom, a 1-inch imaging sensor, 4K video and a 115-degree field of view. On-device computer vision, which Logitech calls RightSight 2, tracks where people are and how they move, driving Grid View, Group View, Speaker View and Presenter View. It is priced at $2,999, available from summer 2026 in graphite and off-white, through Logitech's site and authorised resellers. Two of those specifications matter more than the rest for large rooms. The 1-inch sensor is the meaningful one, because the actual failure mode in boardrooms is not resolution, it is light. Rooms with glass walls, backlit windows or dim conferencing lighting produce noisy, smeared images on the small sensors typical of conferencing cameras, and sensor size is the physical property that addresses it. The dual-camera arrangement is what allows a wide view and a tight view simultaneously, which is what makes speaker framing in a large space work without the constant digital cropping that makes single-camera systems look unstable. The mounting options are a practical detail with more consequence than they appear to have. Ceiling, wall, display and in-wall, with in-wall being Logitech's first, alongside an automatic privacy shutter that indicates when the camera is inactive. Anyone who has specified conferencing hardware knows that the physical install is where projects overrun, and in-wall mounting in a boardroom is an architectural decision rather than an AV one. The part of this product that will change how it is bought is not the camera at all. The Rally AI Cameras detect when, how many and how often people use rooms, and feed that into Logitech Sync for workplace management, including automatically booking and releasing rooms based on people detection. That converts a camera into an occupancy sensor with a management platform behind it, and it moves the purchase decision. A conferencing camera is bought by IT or AV against a per-room budget. A workplace utilisation system is bought by facilities or real estate against a portfolio, and it is justified by property costs rather than by meeting quality. A device that does both can be justified either way, which is a stronger commercial position than the price tag alone suggests. It also introduces a question that buyers should resolve before deployment rather than after. A camera continuously counting people in rooms is generating data about how employees use space and, implicitly, about attendance patterns. The stated purpose is automatic room release, which is a genuine problem: rooms booked and not used are the single largest source of apparent meeting-space shortage in most offices. But the same data supports other uses, and organisations deploying it should decide in advance what is retained, for how long, who can see it, and whether it can be resolved to individuals. In Saudi Arabia and the wider Gulf, personal data protection obligations apply to workplace monitoring, and the sensible sequence is to answer these questions during procurement rather than when someone asks later. On price, $2,999 places the Pro at the upper end of conferencing cameras and well below an integrated multi-camera room system. Against the alternative of a specialist installer building a multi-camera boardroom setup, it is likely to compare favourably. Against a standard single-camera bar for a mid-sized room, it is a considerable step up that only makes sense if the room is genuinely large or the utilisation data is wanted. Several things are not established by any published material. There is no independent assessment of how RightSight 2 handles the cases that break automatic framing: a person standing at a whiteboard partly out of frame, a room where several people speak at once, a long table where the far end is at the edge of the field of view. There is no published latency figure for view switching, which is what makes automatic framing either invisible or distracting. There is no stated regional pricing or availability date for Saudi Arabia or the wider Gulf. And the occupancy detection accuracy in partially occupied or unusually laid-out rooms has not been quantified. The questions a buyer should ask are direct. What is the regional price and lead time. Can I see it running in a room the size of mine, with the lighting I have. What exactly does the occupancy data capture, where is it stored, and can it be turned off while keeping the camera. And what happens to view switching when two people talk over each other, which is what a