"Proctoring software" covers two fundamentally different approaches to the same problem: making sure the person taking an exam is who they say they are, working alone, with only the materials they are allowed to use. Lockdown browsers and consent-based monitoring both try to solve that problem, but they work in almost opposite ways. Understanding the difference matters for anyone evaluating a platform, since the two categories produce very different experiences and very different evidence.
How lockdown browsers work
A lockdown browser is a purpose-built application that takes over the student's device for the duration of an exam. It typically blocks access to other applications, disables copy and paste, prevents switching tabs or windows, and in some cases restricts network access to just the exam platform. The underlying model is control: the software assumes the student cannot be trusted to stay within the rules on their own, so it removes the ability to leave them.
That model has a structural side effect. Anything on the machine that is not part of the sanctioned exam environment, including screen readers, speech-to-text software, or specialized input devices, is competing for the same system-level access the lockdown browser is trying to claim. When the two conflict, the assistive technology usually loses.
How consent-based monitoring works
Consent-based monitoring starts from a different premise: instead of seizing control of the machine, ask the student what they are willing to share, and treat what they share as one stream of evidence for a human to review. In practice, that looks like the same permission prompt a student already understands from video calls. At the start of an assessment, they can choose to share their screen or a specific window through their browser's standard sharing dialog.
Nothing is disabled and nothing is seized. The student keeps their normal browser, their normal accessibility tools, and their normal machine. What changes is that a defined, visible set of streams, which might include webcam video, a 360-degree room view, and optional screen sharing, gets captured and assembled into a single timeline for review. A live indicator shows the student exactly what is being captured at every moment, so nothing happens without their ongoing awareness. You can see this workflow in more detail on the how it works page.
The technical difference: seizing vs. requesting
The clearest way to describe the difference is in terms of who holds control. A lockdown browser operates at the operating-system level, overriding the student's own controls over their machine. A consent-based platform operates at the browser permission level, the same layer that already governs camera, microphone, and screen-share access for any web application. The student grants access explicitly, can see what is active at any time, and the platform never reaches beyond what was granted.
This is not a difference in how much gets monitored. A well-built consent-based platform, using tools like a 360-degree room camera, can actually capture a more complete picture of the exam environment than a lockdown browser plus a standard 78-degree webcam. The difference is entirely in how that capture happens and how much control the student retains while it does.
Why consent-based is gaining traction
Three forces are pushing institutions toward the consent-based model. First, accessibility: because nothing is disabled at the system level, assistive technology keeps working exactly as it normally would. Second, defensibility: evidence captured with explicit, visible consent is far easier for an institution to stand behind in an appeal or a legal challenge than monitoring a student did not fully understand was happening. Third, trust: students respond very differently to a system that asks and shows its work than to one that silently takes over their machine.
None of this requires giving up institutional control. Consent-based platforms still let institutions define exactly what an assessment requires, still flag violations, and still route anything unusual to a human reviewer. What changes is the mechanism: policy is enforced through configured requirements and student consent, not through seizing the machine.
What it means for evidence quality and appeals
The consent-based model also produces a different kind of evidence. Because the student agreed to specific capture at the start of the exam, and could see what was active throughout, there is a clear record of what was and was not being monitored at any given moment. Combined with human review of every flag rather than automatic consequences, that record tends to hold up much better when a student disputes a finding. For a closer look at what separates a strong evidence-based platform from a weaker one, see what to look for in an assessment integrity platform.