**AI Systems Under Siege: Every Organization Targeted in Past Year, Unit 42 Finds**
A new report from Palo Alto Networks’ Unit 42 reveals a stark reality: every organization surveyed has faced at least one attack on its AI systems in the past year. The findings, derived from a survey of over 2,800 participants across 10 countries—including the U.S., UK, Germany, Japan, and India—highlight a growing and systemic vulnerability in AI security, with cloud infrastructure at the heart of the problem.
Conducted between September 29 and October 17, 2025, the research underscores that AI security cannot rely on reactive measures. Instead, organizations must adopt a proactive, scientific approach to safeguarding AI systems, given their complexity and critical applications. The report emphasizes that AI security is inherently tied to cloud infrastructure, where most AI workloads—data storage, model training, and application deployment—reside.
Cloud platforms like AWS, Microsoft Azure, and Google Cloud, while enabling AI scalability, also present prime targets for cyberattacks. Exploitable weaknesses in cloud security can lead to unauthorized access, data theft, or operational disruptions. Traditional security measures often fall short in addressing the unique challenges of AI, such as securing data pipelines, managing identities, and protecting cloud-hosted workloads.
The State of Cloud Security Report 2025 argues that the only effective defense is a holistic approach to cloud security, treating it as foundational to AI protection. This includes enforcing strong policies, encryption standards, regular audits, and isolating AI workloads from cloud vulnerabilities. As AI integrates deeper into sectors like healthcare, finance, and autonomous systems, the stakes rise—breaches could compromise sensitive data, disrupt services, or even endanger lives.
Emerging threats, such as adversarial attacks designed to manipulate AI models, further complicate the landscape. The report calls for collaboration between cloud providers, AI developers, and security teams to build robust frameworks and real-time threat detection tools. The future of AI security hinges on securing the cloud infrastructure that powers it, ensuring resilience against an evolving threat landscape.
Amazon Web Services TPRM report: https://www.rankiteo.com/company/amazon
Palo Alto Networks TPRM report: https://www.rankiteo.com/company/unit42
Google Cloud TPRM report: https://www.rankiteo.com/company/google-cloud
Wakefield Research TPRM report: https://www.rankiteo.com/company/wakefield
"id": "amaunigoowak1766721300",
"linkid": "amazon, unit42, google-cloud, wakefield",
"type": "Vulnerability",
"date": "12/2025",
"severity": "25",
"impact": "1",
"explanation": "Attack without any consequences"{'affected_entities': [{'industry': ['Healthcare',
'Finance',
'Autonomous Vehicles',
'General Enterprise'],
'location': ['Mexico',
'Singapore',
'UK',
'United States',
'Japan',
'India',
'Germany',
'France',
'Brazil',
'Australia'],
'size': 'All sizes (survey included diverse '
'organizations)',
'type': 'Organizations across industries'}],
'attack_vector': 'Cloud infrastructure vulnerabilities, unauthorized access, '
'data pipeline exploitation',
'data_breach': {'data_encryption': 'Recommended but not universally '
'implemented',
'data_exfiltration': 'Possible (if cloud infrastructure is '
'breached)',
'personally_identifiable_information': 'Possible',
'sensitivity_of_data': 'High',
'type_of_data_compromised': ['Sensitive data',
'AI training datasets',
'Personally Identifiable '
'Information (PII)']},
'date_publicly_disclosed': '2025-10-17',
'description': 'Recent findings from Unit 42 (Palo Alto Networks) reveal that '
'every organization has faced at least one attack targeting '
'their AI systems over the past year. The research highlights '
'that AI security is fundamentally a cloud infrastructure '
'issue, requiring a systematic and proactive approach rather '
'than reactive measures. The survey included over 2,800 '
'participants from 10 countries, emphasizing the global scale '
'of the threat.',
'impact': {'brand_reputation_impact': 'Potential erosion of trust in '
'AI-driven services',
'data_compromised': 'Sensitive data, AI training datasets, '
'personally identifiable information',
'identity_theft_risk': 'High (if PII is exposed)',
'operational_impact': 'Disruption of AI-driven services, potential '
'compromise of critical operations',
'systems_affected': 'AI workloads, cloud environments (AWS, '
'Microsoft Azure, Google Cloud)'},
'initial_access_broker': {'high_value_targets': 'AI workloads, cloud '
'environments'},
'investigation_status': 'Ongoing (research findings published)',
'lessons_learned': 'AI security is fundamentally a cloud infrastructure '
'problem. Reactive approaches are insufficient; '
'organizations must adopt proactive, systematic, and '
'scientific methods to secure AI systems. Cloud security '
'must be treated as a foundational element of AI security.',
'motivation': 'Data theft, operational disruption, adversarial attacks on AI '
'models',
'post_incident_analysis': {'corrective_actions': ['Strengthen cloud security '
'policies',
'Implement encryption and '
'identity management best '
'practices',
'Adopt proactive security '
'measures for AI workloads',
'Enhance network '
'segmentation and '
'monitoring'],
'root_causes': ['Weaknesses in cloud security '
'frameworks',
'Insufficient encryption and '
'identity management',
'Lack of proactive security '
'measures for AI systems',
'Over-reliance on reactive '
'security approaches']},
'recommendations': ['Implement strong cloud security policies and encryption '
'standards.',
'Conduct regular security audits of cloud environments '
'hosting AI workloads.',
'Isolate AI workloads from potential vulnerabilities in '
'the cloud.',
'Adopt advanced AI-specific security tools and protocols '
'for real-time threat detection.',
'Collaborate with cloud service providers, AI developers, '
'and security professionals to develop robust security '
'frameworks.',
'Enhance network segmentation and monitoring for AI '
'systems.'],
'references': [{'date_accessed': '2025-10-17',
'source': 'Unit 42 (Palo Alto Networks) and Wakefield '
'Research'},
{'source': 'State of Cloud Security Report 2025'}],
'response': {'enhanced_monitoring': 'Recommended for AI workloads and cloud '
'environments',
'network_segmentation': 'Recommended as part of holistic '
'security approach',
'remediation_measures': 'Proactive cloud security policies, '
'encryption standards, regular security '
'audits, isolation of AI workloads',
'third_party_assistance': 'Unit 42 (Palo Alto Networks)'},
'stakeholder_advisories': 'Organizations are advised to adopt a proactive and '
'scientific approach to AI security, focusing on '
'securing cloud infrastructure as a priority.',
'title': 'Increasing Attacks on AI Systems via Cloud Infrastructure '
'Vulnerabilities',
'type': 'AI System Targeting, Cloud Infrastructure Exploitation',
'vulnerability_exploited': 'Weaknesses in cloud security, insufficient '
'encryption, inadequate identity management, lack '
'of network segmentation'}