As questions around remote testing and test security continue to surface across the assessment landscape, how can programs move beyond perception and focus on evidence? In this episode of Tried and Tested, host Isabelle Gonthier is joined by Paul Gollash, SVP of TOEFL and GRE at ETS, and Wally Dalrymple, Chief Security Officer at PSI and ETS, for a timely conversation on trust, data, and delivery modalities.

Together, they explore how the TOEFL program has evolved a layered security model across both test center and remote delivery. The result is measurable outcomes showing that differences between remote and test center delivery have narrowed to a minimal level. From identity verification and fraud indicators to data driven decision making and continuous improvement, this episode offers a practical and credible look at how security, scale, and trust come together in modern assessment programs.

What you’ll learn:

• Why remote testing and test center delivery have distinct risk profiles, and how purpose built security approaches for each modality strengthen overall test integrity
• How the TOEFL program designed and evolved a layered security model that spans the full assessment lifecycle, from registration and identity verification through delivery, review, and continuous improvement
• What layered security looks like in practice, including how multiple controls work together to deter fraud, detect risk, and protect score integrity when individual signals alone are insufficient
• How ETS and PSI measure the real-world impact of specific security controls, including what the data reveals when individual layers are modified or removed
• Why outcome based measures, such as score patterns and pass rate alignment, provide a more meaningful indicator of trust than incident counts or flagged events alone
• How operating at global scale enables stronger pattern detection, faster response to emerging threats, and continuous refinement of security strategies across programs and modalities
• How AI is influencing both sides of the equation, accelerating new forms of fraud while also strengthening detection, analysis, and decision-making within modern test security programs
• What assessment leaders should consider as trust, security, and delivery models continue to evolve in an environment where risk, technology, and stakeholder expectations are constantly changing

Who should listen:

• Assessment, credentialing, and certification leaders responsible for program integrity and long term trust
• Test security, fraud prevention, and risk professionals designing or evaluating security models across delivery modalities
• Higher education institutions, regulators, and score accepting organizations seeking evidence based perspectives on remote testing outcomes
• Program owners and product leaders balancing access, candidate experience, and rigorous security requirements
• Organizations navigating concern around the credibility of remote English language testing and other high stakes assessments