Raptoric launches AI security testing for high-risk systems under the EU AI Act

Raptoric has introduced a security testing service for high-risk AI systems as companies prepare for EU AI Act compliance. The offering is designed to help organizations prove robustness, cybersecurity and resilience against attacks such as prompt injection, data poisoning and model evasion. Why it matters: - High-risk AI systems under the EU AI Act must meet technical standards for accuracy, robustness and cybersecurity. - The new service targets a growing compliance gap: many AI deployments need testing beyond conventional software security checks. - Organizations also face operational risk if AI systems can be manipulated or degraded after deployment. What happened: - Raptoric launched a security testing service for artificial intelligence systems on June 13, 2026. - The service is aimed at high-risk AI systems covered by the EU AI Act. - Raptoric said the offering helps organizations meet robustness and cybersecurity requirements. - The service covers adversarial testing, robustness testing and cybersecurity testing. The details: - The EU AI Act requires high-risk AI systems to achieve appropriate levels of accuracy, robustness and cybersecurity. - The law also requires high-risk AI systems to be resilient against attempts to manipulate their behavior. - The testing scope includes adversarial testing, data and model integrity review, and resilience testing against prompt injection, data poisoning and model evasion. - Raptoric’s service also reviews training data and model supply chains. - The service assesses the infrastructure around deployed models. - Raptoric provides reporting structured to support AI Act conformity work. - The company said the service is intended to address risks that appear once AI systems are in production. - More information is available at raptoric.com . Between the lines: - The timing matters because EU compliance deadlines for some high-risk AI systems have shifted. - Under the Digital Omnibus on AI, agreed by EU institutions in May 2026 and expected to be formally adopted during 2026, the standalone high-risk AI deadline moves from August 2026 to 2 December 2027. - High-risk AI embedded in regulated products moves to 2 August 2028. - Raptoric is framing the extra time as a chance to build testing into AI development, not as an excuse to delay. - Marko Lukan, Security Consultant at Raptoric, said the engineering challenge did not get easier just because the deadline moved. - Lukan said the companies that use the window to build and document testing will be ready. What’s next: - Organizations with high-risk AI systems will need to decide whether to add security testing now or wait closer to the new deadlines. - The new compliance timetable gives companies more time to document robustness and cybersecurity controls. - Raptoric is positioning its service for firms preparing for EU AI Act conformity work and broader AI operational security needs. The bottom line: - Raptoric is betting that EU AI Act compliance will require security testing as a core engineering step, not a final checklist item.