With increasing levels of competitiveness in national and state level examinations, the need to have stricter levels of surveillance has emerged as one of the key priorities for Examination Boards. Because of the high stakes involved in the examinations conducted by bodies like the CBSE, UPSC, NTA etc, the common question that arises is: Are Traditional CCTV Surveillance Systems Enough to Deter Cheating and Malpractices?
Rajan Verma, VP of Sales - Enterprise and Retail Business at Transline Technologies, believes that while traditional CCTV surveillance is effective to some extent in passively keeping visual records of the examination centres, it most often fails to proactively report incidents of exam malpractices. This is where AI exam monitoring and AI proctoring step in. Using real-time monitoring, facial recognition, expression and body movement tracking, audio analysis and other advanced AI proctoring techniques, examiners can gather accurate live data of attempts of cheating in exams.
According to experts, AI Proctoring is not only about watching, it is more about proactively reporting suspicious behaviour, while addressing AI monitoring concerns in the minds of students and the public in general. Let us have a deeper look at how AI exam monitoring works, how effective it is in deterring cheating in exams and controlling other exam malpractices, and the common concerns being raised regarding it from different quarters. We shall also have a glance at some of the recent applications of AI in exam monitoring in India in recent years.
How does AI Exam Monitoring work?
AI Exam Monitoring goes beyond the static footage of an analog camera. For pen-and-paper exams, AI surveillance systems can be trained with legacy data to identify anomalies in student behaviour. The AI model can actively analyse iris movements of the eye, keystrokes, finger movements, hand and head movements and other behavioral patterns to detect any suspicious behaviour. Moreover, AI surveillance software can analyse voice and audio patterns from audio-capture devices within an examination hall to detect if any candidate is trying to interact with some other candidate or external resources.
AI Proctoring is an even bigger application of AI for online examinations. AI Proctoring softwares use the footage captured on candidate’s device-camera and combine it with microphone audio feed. Then it combines the power facial expression recognition and speech detection to ensure that the candidate is not resorting to exam malpractices, thereby ensuring exam security and sanity. In case of any incident of internet disconnection, screen-switching or detected anomaly, the proctoring software automatically pauses the examination, warns the candidate and records an incident report of the same for the examining authority. Most AI proctoring systems work even with weak and unstable internet connections, and employ strong encryption standards between the candidate’s device and the server.
The most common ones include but are not limited to:
Video feed recording
Facial Recognition and facial pattern analysis
Authentication and verification of candidate
Blocking browser prompts and switching between tabs.
Examination content recording by candidate using mirroring cameras
Noise and Audio-feed detection
Adaptive Testing and Instant Grading
Accurate and real-time exam analytics.
How effective is AI Proctoring and Exam Monitoring?
For pen-paper exams, AI Exam Monitoring has proven to be extremely effective in solving the following use-cases:
Identity verification and Combatting Proxy Test-takers
AI-powered facial recognition systems are more effective and accurate in Identity Verification, and can significantly reduce the number of alleged cases of Proxy Testers taking exams on behalf of candidates in high-stake examinations like NEET, JEE Advanced and SSC CGL.
Lower manpower dependency
AI exam monitoring systems don’t require continuous manual intervention unlike traditional CCTV systems. This frees up critical exam-centre resources for higher-value tasks and minimises associated costs.
Pro-active Alerts vs passive detection
AI exam monitoring solutions are designed to trigger real-time alerts to respective stakeholders like the Central Control Room of the examining authority, floor invigilators and centre heads. Such alerts can be readily acted upon at a faster speed, thwarting any attempts to vandalise exam security.
Exam Security
AI surveillance systems ensure strong levels of exam security by detecting any suspicious activities by nuisance-makers outside examination centers and helping eliminate attempts of mob-vandalism. Moreover, attempts by any candidate to hinder another candidate from answering the exam within the centre premises can be detected by AI algorithms.
Elimination of human bias
Examination AI CCTV systems eliminate human error and bias in proctoring, ensuring consistency across multiple examination centres scattered across geographic locations.
Scalability
AI exam monitoring systems are highly scalable and can be integrated with existing CCTV hardware infrastructure. Data and alerts from thousands of examination centres can be centrally monitored and acted upon in real-time.
What are the Privacy and Legal concerns related to AI exam monitoring?
Despite AI exam monitoring systems displaying high levels of effectiveness and scalability, questions have been raised by different stakeholders over their accuracy and associated privacy concerns. Let delve into these issues and understand how the Government and Examination conduction authorities are navigating these challenges.
Privacy Concerns
The extent to which AI surveillance systems capture and store data has always been a point of concern raised by human rights activists. AI Exam Monitoring softwares do not directly identify individual candidates while proctoring and mostly use aggregate data, thereby minimising privacy concerns.
Data Security Breaches
Unauthorised access to private data captured via exam-centre cameras, data leaks and malware breaches can prove to be detrimental to candidates’ long term career prospects and personal life.
Mental Impact
Continuous examination monitoring for long hours through webcams, exam-room cameras, microphones and other devices are highly likely to make candidates feel uncomfortable and they may perceive it to be an intrusion into their personal spaces. This may cause mental health impacts in the longer run, especially for frequent examination takers.
Bias and discrimination
Several factors like skin colour, iris patterns, appearance and background noise can create discriminatory biases in proctoring. Research studies have further highlighted the need to develop inclusive proctoring solutions and refine AI algorithms to eliminate generic biases.
Algorithmic Transparency and Fairness
Most candidates taking AI-monitored examinations do not have a clear understanding of how these systems work, to what extent data is collected and how the data is stored or used. It is important to spread awareness among candidates regarding these points to ensure fairness and transparency, thereby instilling among them greater trust in these AI systems.
Lack of Specific Legislation
AI Exam Monitoring softwares must comply with regulatory frameworks like GDPR, while also being compliant with Indian laws like Indian Information Technology Act, 2000 (IT Act 2000) and The Information Technology Rules, 2011.
Recent Applications of AI in Exam Monitoring in India
With growing concerns regarding the effectiveness of traditional CCTV surveillance, major examination conduction bodies have started leveraging AI-powered surveillance for critical examinations.
For instance, National Testing Agency (NTA) implemented AI-powered CCTV surveillance to curb exam malpractices. These AI based tools could proactively alert exam centre supervisors in case of any breach of exam security or attempt of cheating by any candidate. Even after the examination, suspicious candidates could be identified by the AI algorithm and acted upon as per the legal provisions.
NTA also conducted the JEE Main Exam 2025 in January across 284 cities across India, using AI exam monitoring. NTA conducted centralised monitoring and stringent identity verification using iFace AI to prevent exam malpractices and ensure exam security across all the centres across the country.
The Union Public Service Commission (UPSC) too has revealed plans to deploy Facial Recognition and AI Surveillance to prevent cheating in exams. As UPSC is responsible for conducting 14 major exams across approximately 80 examination centres countrywide, it is crucial for the Commission to maintain exam integrity.
State Government bodies have also actively started leveraging the power of AI surveillance to combat cheating in exams. In Karnataka, the Class-10 (SSLC) exams were held under the surveillance of AI-powered CCTV cameras in March 2025.
At Transline Technologies, we combine the power of AI-powered surveillance, along with Biometric and Facial Recognition based identity management solutions, to prevent exam malpractices and maintain the highest standards of exam integrity and exam security. Our AI surveillance solutions strongly adhere to the regulatory data security standards and eliminate AI monitoring concerns by addressing common biases.
