Second-screen cheating is becoming one of the most difficult challenges in online exam security. A student can remain inside a locked testing browser while simultaneously consulting a smartphone, tablet, external monitor, AI assistant, or another person outside the controlled exam environment.
That changes the integrity challenge significantly.
Universities, colleges, EdTech providers, certification organizations, and corporate assessment teams can no longer focus only on what happens inside the browser. They also need to consider identity, surrounding devices, learner behavior, unauthorized applications, and abnormal activity throughout the assessment.
As AI cheating in online exams becomes more sophisticated, institutions need security that extends beyond traditional browser restrictions.
Why Browser Lockdown Alone Cannot Address AI-Era Cheating
Browser lockdown remains an important foundation for online assessment security. It can restrict tabs, prevent access to prohibited websites, limit certain keyboard functions, and control applications running on the assessment device.
But its limitation is straightforward: it primarily controls one device.
A candidate may still attempt to access information through another screen, a connected device, external assistance, or an AI-powered tool.
This is where a layered approach becomes important.
Proctortrack addresses this challenge by combining browser-level controls with identity verification, automated or live proctoring, AI-supported monitoring, recorded exam evidence, and review workflows. Its ProctorLock offering, for example, combines automated identity verification with browser lockdown and records video, audio, and desktop screenshots for integrity review.
For higher-risk assessments, Proctortrack also offers automated and live/hybrid proctoring models, allowing institutions to apply different levels of oversight depending on the assessment.
10 AI-Assisted Cheating Risks and How Proctortrack Helps Address Them
1. AI Assistance on a Second Device
A candidate may attempt to consult an AI assistant through a phone or tablet while completing the exam on a secured computer.
Browser lockdown alone cannot directly control an unrelated second device.
Proctortrack helps institutions gain broader visibility by monitoring the testing environment and candidate behavior. Its instructor-facing materials describe monitoring for outside help, unauthorized devices, course materials, and other environmental risks while maintaining identity verification during the assessment.
2. Smartphones and Tablets During Exams
Mobile devices can provide access to search engines, messaging platforms, stored notes, and AI applications.
This makes second-screen cheating an environmental challenge rather than simply a browser problem.
Proctortrack can combine video monitoring, behavioral signals, room-related security checks, and configurable proctoring controls to provide additional context when suspicious activity occurs. Its mobile capabilities also include functions such as mobile lockdown, selfie ID, and room scanning.
3. Unauthorized AI Tools and Chatbots
Generative AI tools can potentially provide explanations, written responses, coding assistance, and other forms of prohibited support.
ProctorTrack’s Pebble secure browser is designed specifically to strengthen the exam environment against unauthorized applications, AI chatbots, search engines, extensions, remote casting, and other applications outside the assessment.
This goes beyond conventional tab blocking by securing more of the desktop and operating-system environment.
4. External Monitors and Screen Casting
A second monitor or casting setup can create another channel through which unauthorized information might be displayed.
This highlights one of the most significant limitations of browser lockdown: controlling a browser does not necessarily mean controlling the entire digital environment.
PEBble includes controls intended to prevent unauthorized remote casting and restrict access to applications outside the secured examination environment.
5. Off-Screen Human Assistance
Not every integrity threat comes from AI.
Another person could potentially provide verbal, written, or visual assistance from outside the main camera’s view.
Proctortrack records and analyzes aspects of the testing session that provide reviewers with greater context for unusual activity. Depending on the configured proctoring level, institutions can combine automated monitoring with live human oversight.
The purpose is not to treat every movement as misconduct, but to provide evidence that can be reviewed when suspicious behavior occurs.
6. Unauthorized Browser Extensions or Applications
AI assistance can also appear through extensions, desktop applications, overlays, or background tools rather than traditional websites.
That makes application-level control increasingly relevant to online exam security.
Proctortrack’s PEBble environment is designed to block unauthorized applications and extensions while restricting access outside the approved assessment environment.
7. Remote Access and Screen Sharing
Remote-control or screen-sharing technologies can introduce another form of outside assistance.
Modern exam integrity solutions therefore need to consider more than web browsing.
Proctortrack’s secure browser capabilities include protections against unauthorized remote casting and applications, adding another layer of security around the assessment workstation.
8. Proxy Test-Taking and Identity Substitution
An exam can be technically secure and still lack integrity if the wrong person is taking it.
That is why identity assurance is a critical part of modern remote assessment.
Proctortrack uses identity-verification capabilities to establish a baseline candidate profile and authenticate exam-session information against it. Its published platform information also describes biometric identity verification and continuous identity-related monitoring throughout testing.
9. AI-Generated Written or Coding Responses
When AI use is prohibited, generative AI can undermine whether an assessment measures the candidate’s own knowledge.
No single AI detector should determine whether academic misconduct has occurred.
A stronger approach combines secure exam delivery, restrictions on unauthorized AI tools, monitoring evidence, clear institutional AI policies, stronger assessment design, and human review.
Proctortrack supports the technology side of this layered model through secured browser environments, automated monitoring, identity controls, and review-oriented proctoring workflows.
10. Suspicious or Abnormal Exam Behavior
Some integrity risks become clearer when multiple behaviors are examined together rather than individually.
Repeated off-screen attention, unexpected environmental events, identity inconsistencies, or unusual activity may warrant closer investigation.
Proctortrack uses computer vision and machine-learning-supported monitoring to help identify policy-compliance concerns at scale and provide results to reviewers in context.
That distinction matters: technology can help identify potentially suspicious activity, while institutions retain responsibility for reviewing evidence and making appropriate academic-integrity decisions.
Proctortrack: A Layered Solution for Modern Online Exam Security
The evolution of cheating explains why online assessment security increasingly requires multiple complementary controls.
Proctortrack brings these controls together through capabilities that include:
- Identity verification: Helps confirm that the registered candidate is taking the assessment and supports continued identity assurance.
- PEBble secure browser: Restricts unauthorized AI chatbots, search engines, applications, browser extensions, and remote casting.
- ProctorLock: Combines automated identity verification with browser lockdown and recorded exam evidence.
- ProctorAuto: Provides fully automated proctoring and integrity reporting for assessments requiring scalable monitoring.
- ProctorLive AI: Combines live certified proctors with automated technology for assessments requiring stronger real-time oversight.
- AI-supported behavioral monitoring: Uses computer vision and machine-learning-based processing to identify policy-compliance signals during assessments.
- Room and mobile security capabilities: Support additional controls such as room scans, selfie identification, and mobile lockdown.
- Integrity review workflows: Provide recorded evidence and contextualized information that instructors can examine rather than relying solely on automated decisions.
This layered structure is important because no single feature addresses every form of AI cheating in online exams.
Browser security protects the device. Identity verification protects candidate authenticity. Environmental and behavioral monitoring provide broader context. Automated systems improve scalability. Live proctoring can provide real-time oversight. Human review supports more defensible decision-making.
Together, these capabilities create a more complete approach to academic integrity in remote learning.
Conclusion
AI has changed both the accessibility and sophistication of unauthorized assistance in online assessments. Second-screen cheating, generative AI, smart devices, identity substitution, and off-screen assistance show why browser lockdown alone cannot provide complete exam protection.
Proctortrack addresses these evolving remote proctoring challenges through a layered approach that combines identity verification, secure browser technology, automated and live proctoring, AI-supported monitoring, environmental controls, and evidence-based review workflows.
For universities, certification providers, EdTech companies, and enterprise assessment teams, the goal should not simply be to detect more cheating. It should be to create assessments whose results remain credible, scalable, secure, and worthy of stakeholder trust.
