Alleged AI-Powered Call Center Breach Exposes Over 10 Million Conversations in the Middle East

An AI-powered call center platform in the Middle East reportedly experienced a significant data breach, allegedly exposing over 10 million conversations between consumers, operators, and AI agents. Attackers allegedly accessed the platform’s management dashboard, stealing sensitive data, including national ID documents. The breach poses reported risks such as phishing, identity theft, and social engineering attacks. The stolen data was reportedly listed for sale on the dark web.

Risk classification

• Primary risk domain: 2 Privacy & Security
• Primary risk subdomain: 2.1 Compromise of privacy by obtaining, leaking or correctly inferring sensitive information

The breach resulted in the unauthorized exposure and theft of non-public sensitive personal data, including national ID documents and private chat logs of over 1 million users.

Additional risk subdomains

• 2.2 AI system security vulnerabilities and attacks: The incident involved a cyberattack where hackers exploited security vulnerabilities to gain unauthorized access to the platform's management dashboard.

Causal factors

• Entity: Human
• Intent: Intentional
• Timing: Post-deployment

The incident was caused by human threat actors intentionally breaching the call center platform's management dashboard after the system was deployed.

EU AI Act risk tier

• Risk tier: 3 Limited Risk

Limited Risk: The report describes an AI-powered conversational call center platform utilizing chatbots, which falls under Risk Level 3 as it involves AI systems interacting with humans.

AI system and alleged parties

• AI system: None named
• AI purpose: Chatbot; Question Answering
• Behaviour type: Assistant
• Alleged developer: Unnamed AI call center platform provider
• Alleged deployer: Cybercriminals
• Alleged harmed parties: Enterprise clients, End users of undisclosed Middle Eastern AI-powered cloud call center platform

Harm severity

Severity ratings are based on this scale.

Highest direct severity in any category: Substantial. Severity is scored from Negligible to Catastrophic in each harm category, for harm the reports describe as caused directly or indirectly by the AI system.

• Physical: direct Negligible, indirect Negligible
• Infrastructure: direct Negligible, indirect Negligible
• Property: direct Negligible, indirect Negligible
• Financial: direct Negligible, indirect Negligible
• Environmental: direct Negligible, indirect Negligible
• Malicious content: direct Negligible, indirect Negligible
• Differential treatment: direct Negligible, indirect Negligible
• Civil rights: direct Negligible, indirect Negligible
• Democracy: direct Negligible, indirect Negligible
• Privacy: direct Substantial, indirect Negligible
• Psychological: direct Negligible, indirect Negligible
• Epistemic: direct Negligible, indirect Negligible
• Child sexual exploitation and abuse: direct Negligible, indirect Negligible

Privacy

Reported: The report explicitly describes massive privacy violations where over 10 million conversations containing PII and national ID documents were stolen and offered for sale.

Directly caused: The breach directly exposed the personal data, chat logs, and national ID documents of approximately 1 million end users.

Inferred additional harm: It is highly likely that the exposed PII will be exploited for identity theft, leading to further ongoing privacy violations and security compromises for the affected individuals.

People affected

• Occurrences reported: 1
• People reportedly harmed: 1000000
• People reportedly exposed: 1000000

Potential causes

Management

• Poor AI Supply Chain Control: Management absorbed third-party AI risks without proper governance.
• Inadequate Data Retention Policy: Management failed to enforce strict limits on retaining customer chat logs.

Technology

• Insecure Management Dashboard: Attackers gained unauthorized access to the platform's control dashboard.
• Vulnerable API Token Management: Token manipulation allowed malicious injection into integration channels.
• Black Box AI Architecture: Lack of transparency regarding data protection and retention in AI systems.

Data Inputs

• Unsanitized PII Retention: AI system retained national IDs and sensitive customer data in chat logs.
• Lack of Input Minimization: Users uploaded sensitive files directly to the conversational AI bot.

Human Factors

• Compromised Engineer VPN: Attackers stole VPN and admin credentials belonging to an engineer.
• Unwarranted Trust in AI Agent: Users assumed the AI session was secure, exposing them to hijacking.

Process and Methods

• Inadequate AI Risk Assessment: Systems were deployed without comprehensive AI supply chain risk assessments.
• Lack of Zero-Trust Mindset: Failure to assume breaches would occur in third-party hosted AI systems.

Regulatory Environment

• Weak AI Data Governance Rules: Regulatory frameworks did not enforce strict PII sanitization in AI models.

Information quality

• Classification confidence: High
• Reason for confidence: The reports provide consistent and detailed information regarding the nature of the data breach, the volume of compromised records, and the specific types of sensitive data exposed.
• Ambiguities identified: The specific name of the call center platform and the exact technical vulnerability exploited to access the dashboard are not disclosed.
• Alternative interpretations: The incident could be viewed purely as a traditional SaaS database breach, but the central role of conversational AI in gathering and retaining the sensitive chat logs makes it an AI safety and privacy incident.

A cyber breach of a Saudi Arabian AI-powered call center platform resulted in the theft of 10 million chat logs and national ID documents of 1 million users. While the incident represents a significant privacy failure and raises the risk of future identity theft and social engineering, its direct national security impact remains minor and manageable under standard cybersecurity protocols.

• Overall national security impact: Minor
• Scope: Single nation
• Primary target: Saudi Arabia
• Alleged perpetrator: Unknown

Threat characteristics

• Imminence: Near-term. The database has been leaked and put up for sale, posing a near-term threat of phishing and identity fraud to the affected individuals.
• Autonomy: Human-controlled. The AI system functioned as a standard customer service chatbot, and the breach itself was executed by human threat actors targeting a database.
• Novelty: Established threat. Data breaches targeting cloud databases and customer management dashboards are well-established cyber threats.

Impact by dimension

• Physical security: Negligible. No threat to physical systems, kinetic assets, or critical infrastructure was reported in connection with this data breach.
• Information security: Minor. Stolen conversational data and PII could facilitate targeted social engineering or phishing, but there is no evidence of active state-sponsored information warfare.
• Sovereignty: Minor. Compromise of national ID documents of citizens poses a minor risk to identity systems, but does not disrupt core government operations or decision-making.
• Economic security: Minor. While 1,000 enterprise customers were affected and data was sold on the dark web, there was no theft of critical AI models, intellectual property, or systemic financial impact.
• Societal stability: Minor. The breach represents a significant privacy violation for over 1 million users, but does not constitute systematic state surveillance or algorithmic oppression.