NCBE Completes Its First Remotely Proctored Pilot Test
By Sabrina Dubbert
On August 29, 2026, NCBE administered its first remotely proctored pilot test for the NextGen Uniform Bar Examination. Fifty-six research participants completed one full section of exam content from their individual testing locations. To proctor the test, NCBE utilized a combination of automated monitoring and human oversight, intervention, and review through the ProctorNow system by Internet Testing Systems (ITS).
This research study was designed to assess the feasibility of remote proctoring for pilot testing purposes only. NCBE is not exploring remote administration of the operational bar exam. With this study, our organization sought to answer several questions:
- Is remote proctoring suitable for future pilot tests?
- Will exam content remain secure in a remote testing environment?
- Is remote proctoring operationally feasible?
- What is the candidate experience?
Prior to the pilot test, participants were required to download the operational secure browser on their testing devices and the Remote Proctor Connect app on their phones; the latter could serve as a second camera. They were asked to clear their testing area of books, source materials, and unauthorized devices, and nonparticipants were not permitted to enter the testing location.
The pilot test began at 9:00 a.m. in each participant’s respective time zone. During the check-in process, the ProctorNow system verified the participant’s identity and checked web cameras, microphones, the network connection, and the testing area. Once testing began, participants were given three hours to complete 40 multiple-choice questions, two integrated question sets, and one performance task. NCBE staff monitored the pilot test, and ITS was available to provide test-day support.
Potential Benefits of Remote Proctoring for Research Purposes
NCBE continually works to develop new questions for future bar exam administrations. To deepen the question pool, we must be able to pilot more content, more frequently. In-person pilot tests remain a critical part of this ongoing development process. At the same time, remote proctoring presents another option—one that could broaden the participant pool, test content more often, and create a more efficient research cycle.
A wider pool of participants is essential for pilot testing to more accurately represent the candidate population taking the operational bar exam. Remote proctoring allows volunteers to participate from home, eliminating geographical constraints and the need to travel for an in-person event.
In addition to broadening the participant pool, remote proctoring would allow NCBE to test exam content more frequently because there’s no need to wait until the next scheduled in-person event. This testing mode also makes it possible to recruit participants on a rolling basis rather than focusing recruitment efforts around in-person events.
Finally, remote proctoring could shorten the research cycle, allowing us to more quickly collect data, analyze it, and act upon research outcomes. The more efficient our research process, the sooner we can pursue initiatives that improve the bar exam. Those initiatives are critical to protecting the public and the integrity of high-stakes licensure exams.
ProctorNow: Automated Monitoring with Human Oversight
NCBE’s latest research study was designed to assess the validity of remote proctoring itself. However, we also sought to determine whether the ProctorNow system is an effective, reliable platform for future pilot tests.
ProctorNow has the key advantage of combining automated monitoring with human oversight. AI technology reduces the need for live proctoring while preserving content security and the validity of exam results. ProctorNow can screen for prohibited events, such as the participant leaving camera view, other people entering the testing area, or the presence of unauthorized devices or source materials. If the AI tool detects evidence of a concerning activity, it will notify a live proctor, who can choose to intervene or flag the event for post-test review.
Although automated security features are available, human oversight, intervention, and review remain a critical part of the ProctorNow system. If a candidate needs assistance during the test, they can use a live chat feature to speak with ITS support staff. Once pilot testing is complete, human proctors review all video recordings, including any footage flagged for concerning activity. Proctors also decide whether to escalate potential security concerns for additional review.
It is important to note that the AI tool does not decide when cheating occurs. Under the ProctorNow system, human proctors always review relevant footage, assess the situation, and determine if escalation is necessary. The AI tool simply monitors events and notifies a live proctor when certain event criteria are met. Humans make the final call.
What Did We Learn?
Preliminary data are encouraging. All 56 participants were able to download the secure browser and complete the security check-in process. On average, the ProctorNow system only took a few minutes to check the technology on each testing device.
Throughout the pilot test, the system maintained good communication between examinees and live support staff. Very few technology-related issues occurred, and they were resolved quickly. These issues were specific to individual candidate devices; none relating to the ProctorNow system were detected.
All 56 examinees successfully completed the full section of pilot content, and all responses were successfully collected and submitted through the testing platform.
One of the main objectives was to assess content security through the ProctorNow system. To achieve this objective, one NCBE staff member served as a “plant” to intentionally introduce prohibited behavior. The ProctorNow system successfully flagged this behavior and notified a live proctor, who then implemented the appropriate escalation measures. During the test, no true examinees were observed attempting to copy or reproduce content. Nevertheless, NCBE limited the consequences of potential security incidents by using questions that will not be funneled into the operational pipeline for actual bar exam administrations.
These initial results appear promising, but more research is needed to determine whether remote proctoring is appropriate for pilot testing exam content. The full report of this research study will become publicly available at a later date.
Conclusion
This pilot test was the first step toward a decision about remote proctoring for research purposes. The test was intentionally narrow and not intended to evaluate the full range of candidate configurations. If NCBE continues to explore remote proctoring, future studies will include accommodations and other accessibility concerns to help determine whether remote proctoring technology remains feasible for piloting exam content.
No matter the pilot testing environment, our mission remains the same: to conduct ongoing research that continually supports high-stakes licensure and the legal profession.
Sabrina Dubbert is Editorial Assistant for the National Conference of Bar Examiners.