Proctorio accused of racial bias in AI proctoring during online exams
Multiple news outlets reported in mid to late 2020 that Proctorio’s AI-based remote proctoring and facial-recognition tools were alleged to have discriminated against students, particularly students of color. Coverage and campus protests raised questions about biased detection and identity-verification failures in automated proctoring systems.
Automated facial-recognition and behavior-detection in proctoring allegedly misidentified and penalized students of color.
Key facts
- What
- Multiple news outlets reported in mid to late 2020 that Proctorio’s AI-based remote proctoring and facial-recognition tools were alleged to have discriminated against students, particularly students of color.
- Incident date
- Nov 7, 2020
- Who
- Proctorio
- Failure mode
- Policy Violation
- AI surface
- Computer Vision
- Severity
- High
What happened
In November 2020 and earlier in the pandemic several news outlets and student groups reported allegations that Proctorio’s remote proctoring software and its automated facial-recognition and behavior-detection features disproportionately flagged or failed to verify students of color. Reports described students being subjected to additional scrutiny, failing verification checks, or experiencing stress and disruption during exams. These accounts were framed as allegations in media coverage rather than adjudicated findings.
What broke inside the model
- 01 · TriggerA prompt pushes against a deployment boundary.
- 02 · Model stepThe model produces the disallowed output.
- 03 · Control gapNo enforcement blocks it at generation time.
- 04 · FailureThe output crosses the policy line.
- 05 · ConsequenceA limit the business set is breached in public.
The output crosses a policy boundary the deployment had defined.
Reports allege failures in automated identity-verification and behavior-detection components, specifically that facial-recognition or related models performed worse for darker-skinned students and therefore produced false flags or verification failures. The mechanism described in press coverage points to biased model performance and insufficient demographic validation during deployment, rather than human-intentional discrimination.
What it cost
Sources
- PressRemote testing monitored by AI is failing the students it is supposed to helpnbcnews.com
- PressOnline exams raise concerns of racial bias in facial recognitioncsmonitor.com
- PressSoftware that monitors students during tests perpetuates inequality and surveillancetechnologyreview.com
Cite this entry
https://failureindex.ai/failures/proctorio-accused-racial-bias-proctoring-duringAI Failure Index. "Proctorio accused of racial bias in AI proctoring during online exams" (FI-0348). Realm Labs. https://failureindex.ai/failures/proctorio-accused-racial-bias-proctoring-during (indexed Jun 9, 2026).Data fields CC-BY 4.0, prose citation permitted. Incident ID FI-0348. Full dataset at /data.
Note from Realm Labs, the Index steward
How Realm fits
- Prism
- OmniGuard
This entry sits in the index's predictive wing: a system that scores, ranks, perceives, or steers rather than generates. Realm's runtime layer is built for the generative and agentic systems now moving into these same decision seats, where it watches a model's internal state and holds an unsupported claim or an unchecked action before it commits. The control gap on this record, an automated decision that reached people with no runtime check in front of it, is the same gap. The index keeps predictive failures on the record because the pattern carries straight into the systems shipping today.