Abnormal AI has launched Attune 1.0, a behavioral foundation model designed to help organizations defend against the growing wave of AI-driven cyberattacks. With this release, the company is strengthening its AI-native behavioral security platform and giving enterprises a more unified way to detect advanced threats that no longer follow predictable patterns.

Organizations rely on communication to build trust across employees, partners, and customers. However, attackers are now using AI to exploit that trust at scale. They are creating highly personalized campaigns that look authentic, feel context-aware, and are harder to detect with conventional defenses. As a result, security teams can no longer depend only on static rules or known threat intelligence. Instead, they must prepare for a world where every malicious message could appear unique.

Abnormal AI says Attune 1.0 answers that challenge by combining the company’s eight years of behavioral insight into a single model. Trained on more than one billion derived behavioral signals, the model now powers 85% of detections across the Abnormal Behavior Platform. In addition, it creates a shared intelligence layer that supports the company’s broader and expanding security portfolio.

“Attackers are leveraging AI to imitate trusted behavior so convincingly that static rules and threat feeds struggle in the era of AI-driven attacks,” said Evan Reiser, CEO and Co-Founder of Abnormal AI. “Attune 1.0 is how we close that gap—with a behavioral foundation model that understands normal organizational communication patterns. It gives customers a single intelligence layer that understands known good behavior, catches what isn’t, and strengthens every product we ship as part of the Abnormal Behavior Platform.”

Unlike older security systems that treated identity, behavior, and content as separate inputs, Attune 1.0 uses a unified multimodal architecture. Because it learns these signals together, the model can better understand when different signals reinforce one another and when they conflict. Consequently, it becomes easier to reveal suspicious patterns that attackers intentionally try to hide. From the beginning, Abnormal has focused on understanding what normal behavior looks like inside a customer’s environment. That approach now allows the platform to detect AI-driven attacks when behavior suddenly deviates from trusted patterns.

The company also shared several important milestones tied to Attune 1.0. According to Abnormal AI, the model is already detecting approximately 150,000 more attack campaigns per week than earlier systems. This improvement highlights its ability to catch sophisticated messages that previously went undetected. At the same time, Attune delivers 50% higher precision, which helps reduce false positives and improves the overall efficiency of security teams.

Another notable example shows the model’s ability to identify emerging threats before they become widely known. Abnormal said Attune recently detected and blocked a novel Microsoft Teams OAuth phishing campaign two months before it was publicly documented. That kind of early detection is especially valuable in today’s threat landscape, where attackers constantly test new methods to bypass traditional defenses.

Beyond detection, Abnormal AI is also expanding visibility and control for cloud email security. With Detection 360 Insights, now generally available, analysts can see the behavioral reasoning behind every AI-driven determination. This added transparency helps teams understand exactly why a message was flagged. Meanwhile, Custom AI Models, currently in early access, allow security teams to guide the AI using natural language descriptions tailored to their own environment. Therefore, customers can fine-tune protections based on business-specific communication patterns.

Abnormal is also enhancing human risk management through updates to its AI Phishing Coach. Instead of relying on generic compliance training, the company is moving toward a more automated and personalized model. New features such as Phishing Risk Scoring provide continuously updated readiness signals based on simulation activity, reporting behavior, and training outcomes. In addition, BEC and VEC Simulations now mirror interactions involving managers, coworkers, and vendors by using data from the Abnormal relationship graph.

Overall, Attune 1.0 marks a significant step in Abnormal AI’s effort to help enterprises secure modern communication. By combining behavioral intelligence, stronger detection, higher precision, and personalized human risk training, the company is giving organizations a more adaptive way to fight back against AI-driven attacks.

Recommended Cyber Technology News:

To participate in our interviews, please write to our CyberTech Media Room at info@intentamplify.com