Assessment Fraud Doubles in 2025, Entry-Level Hiring Most Vulnerable

  • Cheating and fraud attempt rates for proctored assessments more than doubled in 2025, rising from 16% to 35%.
  • Entry-level assessment cheating and fraud attempt rates nearly tripled year-over-year, jumping from 15% to 40%.
  • CodeSignal's detection system flagged 35% of cheating attempts involving frequent off-screen referencing, 23% with unusually linear typing patterns, and 15% with elevated similarity to known answers.
  • Cheating and fraud attempt rates reached 48% in Asia-Pacific compared to 27% in North America.

The sharp increase in cheating and fraud attempts highlights the growing challenge of maintaining hiring integrity in an era of widespread AI tool adoption. CodeSignal's data underscores the critical role of robust fraud prevention infrastructure, particularly for entry-level roles where the risk is most concentrated. The regional disparities in fraud attempt rates suggest that cultural and technological factors may be influencing the prevalence of cheating behaviors.

Fraud Evolution
How the sophistication of cheating methods will evolve as AI tools become more integrated into daily life.
Regional Disparities
Whether the significant regional differences in fraud attempt rates will persist or narrow over time.
Entry-Level Risk
The pace at which entry-level hiring will adapt to the heightened risk of fraud and cheating.
AI Hiring's Integrity Crisis: Cheating Attempts Doubled in 2025