SentinelOne is utilizing OpenAI Daybreak to transform its malware analysis operations, reducing a process that previously took top security researchers four to six weeks down to roughly eight hours. According to Gregor Stewart, Chief AI Officer at SentinelOne, the tool produces human-level reverse-engineering output, enabling the company to investigate a significantly larger volume of threats. Historically, deep investigations were restricted to a small number of high-priority samples due to time constraints, which limited the ability to connect findings across different attacks. With faster turnaround times, researchers can now compare numerous software samples, spot shared components, and identify broader relationships between threats to build more proactive defenses. Furthermore, access to OpenAI cyber capabilities has democratized security work within SentinelOne, allowing a wider range of team members to conduct investigations rather than relying solely on a small group of specialists. Stewart notes that while automation handles point fixes and routine analysis, human accountability remains mandatory, especially for major architectural decisions where human oversight is still required.
SentinelOne cuts malware analysis from weeks to hours
SentinelOne is using OpenAI Daybreak to reduce the time required for complex malware reverse engineering from weeks to approximately eight hours. This acceleration allows security researchers to investigate a higher volume of threats and uncover relationships between previously isolated software samples. While automation speeds up analysis, SentinelOne maintains human accountability for critical decisions.
Why it matters
This application demonstrates how domain-specific AI tools can dramatically compress complex technical workflows and scale high-value expertise across an enterprise. It highlights the practical balance between automated efficiency and human oversight in high-stakes cybersecurity operations.
Topic dossiers OpenAI