Insider AI Threat

Insider AI Threat

Securing Enterprise from Insider Threat: AI Agents and Stolen Identities

Crown Melbourne, Southbank, Australia

Hosted by Factor

Overview

For many organisations, the most dangerous threats now originate from inside the network — not because employees are malicious, but because attackers increasingly exploit legitimate identities . Compromised credentials, automated scripts, service accounts and emerging AI agents can all operate with valid access, making malicious activity difficult to detect with traditional rule-based security tools.

The challenge for security leaders is that these threats often appear indistinguishable from normal behaviour. A compromised account accessing sensitive systems or an AI-driven workflow pulling unexpected data may generate no obvious alerts.

This roundtable will explore how behavioural analytics and identity-centric detection can uncover these hidden threats by identifying subtle deviations from normal user and system activity. It will examine how organisations can strengthen insider threat programs while maintaining visibility across both human and machine identities.

Discussion Topics

Insiders Without Intent: The Rise of Identity-Based Threats

Why compromised credentials now underpin many major breaches

The difference between malicious insiders, negligent users, and hijacked accounts

When Automation Becomes the New Insider

AI agents, service accounts and automated workflows accessing enterprise systems

The growing challenge of monitoring non-human identities

Behavioural Analytics and UEBA: Detecting the Subtle Signals

Establishing behavioural baselines for users, devices and accounts

Detecting anomalies such as unusual access patterns, impossible travel or abnormal data movement

Balancing Detection with Trust

Strengthening insider threat detection without creating a culture of surveillance

Governance, transparency and privacy considerations

Agenda

Partners

More events from this host