The MITRE ATT&CK Enterprise Evaluations, often called the Olympics of cybersecurity, released results for Enterprise Round 7 (ER7) in December 2025, revealing that the best-performing vendors blocked only 31% of adversarial actions. The evaluation, the most demanding in the program's history, included cloud adversary emulation, identity-centric attacks, and cross-environment lateral movement for the first time simultaneously. Among the nine vendors that participated, CrowdStrike and Cybereason tied for the highest protection score, leaving 69% of attacks unblocked. Notably, all nine vendors scored zero percent on identity attack blocking, targeting techniques used by Scattered Spider, the criminal collective behind the MGM Resorts and Caesars Entertainment breaches. Cloud attack blocking rates ranged from 0% to 7.7%, with five vendors blocking nothing in the first AWS adversary emulation.
Three of the largest cybersecurity vendors—Microsoft, SentinelOne, and Palo Alto Networks—withdrew from ER7, citing various reasons. Microsoft pointed to its Secure Future Initiative, SentinelOne described the evaluations as "PR-driven," and Palo Alto Networks cited internal innovation focus. This withdrawal trend reflects a 63% decline in participation from 30 vendors in 2022 to 11 in 2025.
VectorCertain LLC, a smaller player, took the opposite approach. Using MITRE's published ER7 adversary emulations—Scattered Spider and Mustang Panda—VectorCertain ran its SecureAgent platform through a rigorous self-evaluation spanning Sprints 30–34, completed February–March 2026. The company extended the evaluation to include Volt Typhoon, a third adversary targeting U.S. critical infrastructure, and added behavioral governance and memory governance testing. Results showed 14,208 tests executed with zero failures, a 100% protection rate against all three adversaries, and zero false positives. VectorCertain has formally enrolled in MITRE's Enterprise Round 8 (ER8) for independent verification.
The architectural difference behind SecureAgent's performance is a four-gate governance pipeline that evaluates every proposed AI agent action before execution, rather than detecting threats after they occur. This approach addresses identity abuse, which does not generate endpoint telemetry, by governing actions at the point of intent using policy, not signatures.
The ER7 results have macroeconomic implications. Global fraud and cybersecurity losses totaled $485.6 billion in 2023, according to Nasdaq Verafin, and AI-specific cyberattacks cost an estimated $15 billion in 2024. TransUnion reported companies lose 7.7% of annual revenue on average to fraud, with U.S. figures reaching 9.8%. IBM's 2025 Cost of a Data Breach Report found that organizations deploying AI in prevention workflows saved an average of $2.22 million per breach.
VectorCertain's SecureAgent enters ER8 as the first AI Safety and Governance platform in the program's history. The company's founder, Joseph P. Conroy, has a 25-year history of building AI systems for critical applications, including industrial gas detection and predictive diagnostics. For more information, visit vectorcertain.com.
