Six years ago, in a pre-pandemic world, few of us would have predicted the scale and prominence of remote proctoring across the assessment sector. What began as an emergency sticking plaster has become an integral part of high-stakes assessment and professional certification.
COVID forced many of us away from what felt, at the time, like the safety of test-centre-only delivery and into remote models at unprecedented speed. Candidates and clients quickly adapted to the idea of sitting assessments at home, and blended delivery models combining centre-based and remote options have since become common. Remote proctoring has unlocked access for candidates who live far from centres, have mobility or caring responsibilities, or need more flexible scheduling.
Yet despite this rapid normalisation, many organisations are still grappling with a fundamental question: what does “good” look like?
We are balancing candidate experience, fairness, security, operational capacity and regulatory scrutiny in an environment that is once again being reshaped, this time by AI-assisted cheating and rising expectations of flexibility and access. Remote proctoring is no longer an emergency workaround. It is a strategic decision with long-term implications.
In this series, we will explore eight themes that together determine whether remote proctoring earns and keeps trust:
- Fairness across delivery channels
- Reviewer workload and flag fatigue
- How Secure Is Secure Enough?
- Choosing the right proctoring model
- Candidate experience and wellbeing
- Equity, access and global reach
- Governance, policy and transparency
- AI, human judgement and the future of trust
This opening article sketches each theme in brief. Over the coming weeks, we will return to each one in depth.
1. Fairness across delivery channels
When candidates can choose between a test centre and a remotely proctored exam, are the conditions truly comparable?
Variations in device quality, bandwidth stability, room environment, household interruptions and digital confidence all influence performance. If one route is perceived as easier, or more aggressively policed, we risk creating a process that undermines confidence in results. The risk is not technical failure. It is the emergence of a two-tier system.
Thoughtful organisations are therefore asking not whether remote delivery is possible, but how policies, support and controls can ensure equivalence and defensibility across channels.
2. Reviewer workload and flag fatigue
The operational engine room of remote proctoring is rarely visible, yet it carries significant weight. Behind every remotely delivered exam is a team who are diligently reviewing flags, video clips and incident reports. In many organisations, the review stage now represents a significant part of the post-assessment workload.
AI-generated alerts, particularly where thresholds are poorly calibrated, can create an overwhelming volume of potential incidents. Without careful design, this leads to “review fatigue”: reviewers become overwhelmed, tired, decision-making slows, and the consistency of integrity judgements suffers.
The hidden human cost is significant, and often underestimated. Staff wellbeing and judgement quality are directly linked to system design. Smarter triage, clearer thresholds and stronger reviewer training are not operational luxuries. They are integrity safeguards, and they deserve an article of their own.
3. How Secure Is Secure Enough?
Proctoring platforms have evolved significantly in the years that have followed the pandemic: multi-factor ID verification, browser lockdown, multi-camera options and AI behaviour analysis are now standard features.
Yet this evolution has coincided with increasingly sophisticated attempts to bypass controls. Generative AI, organised collusion and real-time external assistance tools have intensified what can feel like a perpetual arms race. Recent decisions by prominent professional bodies to step back from remote exams for certain cohorts have encouraged many to voice fears they had previously held privately: are current approaches sufficient for the highest-stakes assessments?
That question deserves a better answer than simply adding more controls. Security and validity are inseparable. Too little security, and a result stops telling us what a candidate knows. Too much, and we begin measuring composure under surveillance rather than the capability we set out to assess. There is no universal definition of “secure enough”. The answer depends on what the task exposes, what is at stake if the result is wrong, and the scale we are working at.
4. Choosing the right proctoring model
Every qualification we deliver is unique, and there is no one-size-fits-all solution. As proctoring platforms have evolved, it is more important than ever that we understand exactly how each model operates.
The right mix of in-centre, live remote, AI-assisted or hybrid proctoring depends on the assessment purpose, the stakes, the candidate profile, the geography, even the regulatory environment we work in. It also depends on our organisation’s attitude towards risk. For some entry-level or low-stakes assessments, lighter-touch online invigilation may be appropriate. For higher-stakes professional qualifications, the challenges and the consequences are greater, and many are now considering a predominantly centre-based model, supplemented by tightly controlled remote options where necessary.
The real challenge is not allowing technology, or fear of technology, to drive strategy. Instead, we should be moving towards governance-driven decisions: seriously considering our objectives and our risks, then implementing a model that can meet both.
5. Candidate experience and wellbeing
On the surface, remote proctoring looks like a candidate’s dream: no travel, familiar surroundings, fewer logistical uncertainties.
The reality can be very different. Technical failures can leave candidates feeling isolated and anxious. Those less confident with technology may find the process stressful. Some worry about connectivity issues, or being falsely flagged because of household interruptions, a family member bringing in tea or a cat unplugging the power cable. Others feel uneasy about being monitored in their own home.
In-centre exams bring their own stresses: the time and cost of travel, an unfamiliar exam hall that can feel overwhelming, and accessibility limitations in some locations. A mature delivery strategy must therefore consider not only security, but the psychological and practical realities faced by candidates, particularly if we are genuinely committed to widening participation and supporting more diverse cohorts.
6. Equity, access and global reach
One of the strongest arguments for remote proctoring has always been equity. For candidates who live in rural areas, in locations without an established test-centre network, or with specific access needs, remote delivery can be the difference between participation and exclusion.
However, equity depends on reliable connectivity, appropriate devices and access to a suitable assessment environment, and none of these can be assumed. And if organisations were to retreat entirely from remote options, access may narrow again.
The equity question is therefore no longer simply “can we deliver remote assessments?” It is “how do we maintain access within a defensible remote model?”
7. Governance, policy and transparency
Technology alone cannot guarantee assessment integrity, nor should we expect it to.
The way our organisations set and apply policies, from incident classification and sanctions to appeals processes, data retention and the role of AI in decision-making, now plays a central role in how fair and reliable our assessments appear. It sits at the very centre of our organisation’s defensibility.
Our candidates rightly expect more from us than ever. What we do is critical to their futures. They expect clear, transparent explanations of what will be monitored, how flags will be handled and how they can challenge decisions. They expect consistency across in-centre and remote assessments. Regulators and clients, meanwhile, demand proportionate, consistent, evidence-based security controls. Good governance transforms uncertainty into credibility and suspicion into defensible confidence.
8. AI, human judgement and the future of trust
Looking ahead, I believe the debate will move on from “remote versus centre” towards a better question: how can we best blend human and digital capabilities fairly, ethically and responsibly?
AI will continue to expand its role in anomaly detection, identity verification and post-assessment analytics. But it cannot replace human judgement, particularly in borderline cases, appeals and decisions where a candidate’s future depends on the outcome. AI can become a valuable player in the team, but it must remain part of a team that is led by human judgement.
The future lies not in automation replacing expertise, but in clearly defined collaboration between algorithmic detection and human interpretation.
What Happens Next?
Remote proctoring is no longer a temporary solution or a technological experiment. It is a structural component of modern assessment delivery.
The organisations that navigate this space successfully will be those that approach it not as a product decision, but as a governance challenge: openly balancing access, integrity, operational reality and trust.
Each of the eight themes above is pivotal to the future role of proctoring in assessment, and each deserves a deeper exploration in its own right. In the weeks that follow, we will look at every one in detail.
In a debate often characterised by polarisation, the aim is not to advocate for or against remote delivery, but to examine how it can be designed, thoughtfully, transparently and proportionately, within a complex and rapidly evolving environment.
