There is a trend in educational technology right now: the idea that the best way to prevent cheating is to treat every student as a suspect.
There are platforms that demand 360-degree room scans, track eye movements, log every keystroke and record audio, all in the name of integrity. There is a cost to that surveillance which rarely appears on the invoice. It creates anxiety, it runs over any reasonable expectation of privacy, and it reads to a student as a statement about what you think of them.
When we started building this, we asked a different question: what if the route to better academic integrity is collecting less, not more?
The privacy-anxiety loop
Here is something that does not get said often enough: when students feel watched too closely, their performance drops.
It is not only discomfort, it is cognitive load. A student sitting an exam should be thinking about the question in front of them, not wondering whether looking down at the desk will trigger a flag, or whether a shadow moving across the room will be read as suspicious.
Anxiety is the enemy of good assessment. When part of a student's attention is spent monitoring the monitoring, they are not doing their best work, and the exam stops measuring the thing it was written to measure.
A platform that respects privacy is not being nice. It is creating the conditions in which a student can actually show you what they know.
Only what we need
Conventional wisdom says more data equals more security. We disagree.
We collect only what is needed to run the exam, know whose paper it is, and return the result. A name, an email address so the result can reach them, their answers, and whether the attempt was left. That is the list.
To be specific about what we do not do:
- We do not track eye movements.
- We do not record the room or ask for a 360-degree scan.
- We do not harvest behavioural data for analytics or profiling.
- We do not sell student information to anyone.
- We do not use student exam responses to train AI models.
This is not minimalism for its own sake. Every piece of data you collect is a responsibility you have taken on, and every invasive feature is one more thing standing between a student and their best work.
When we say the exam is secure, we mean the attempt locks if the exam screen is left and the answer key is never sent to the device. We do not mean we have turned a student's phone into a surveillance instrument.
Trust as an integrity control
Here is the counterintuitive part: when you treat students as adults and explain the rules plainly, they are more likely to follow them.
Think about what happens when an educator can say, "the exam locks if you leave the screen, and that is the whole of it. We are not recording your room or watching your eyes." The student understands the boundary. The rule is legible and proportionate. They are far more likely to accept it.
Compliance goes up when the rules make sense. This is what we mean when we say integrity is partly a communication problem. Students should be told, in plain language, what is restricted, what is monitored, why those controls exist and how their privacy is handled.
No legalese. Nothing important buried on page four of a terms document.
When students understand the reason behind a control, they stop reading it as an arbitrary restriction and start reading it as protection for the value of their own work. That is where real academic integrity comes from: not the enforced kind, the willing kind.
What this adds up to
We built this because assessment should measure learning rather than technical patience. It should give educators useful evidence instead of paperwork. And it should produce a result everyone can believe, under conditions everyone understood in advance.
Collecting less data is not a compromise on security. It is a different theory of it, one that treats trust as a feature rather than a gap to be closed.