The Acceleration of Cloud Attacks and AI Adoption
The speed at which attackers move from initial access to data exfiltration in the cloud is increasing, challenging traditional defense models. Palo Alto Networks’ latest State of Cloud Security report, now in its fifth iteration, surveyed 2,800 respondents across 10 countries and found that artificial intelligence tools are reshaping cloud environments faster than organizations can secure them. Software supply chain attacks, including malicious packages in open source repositories and backdoors in widely used libraries, continue to grow in frequency and impact, threatening not only security teams but the broader business and its customers.
The Rise of Digital Colleagues and Invisible AI Agents
Enterprises claim readiness to secure AI, but most lack the tools to detect or control AI infrastructure. Delinea CEO Art Gilliland warns that relaxed governance and invisible AI agents are creating serious enterprise risk. As organizations shift from blocking AI to building with it, they are effectively onboarding a new class of “digital colleagues” operating at machine speed. Goldman Sachs estimates that agentic AI could account for 60% of the software market. Meanwhile, credential based attacks are accelerating as AI enhances phishing and exploitation tactics, with Dashlane CEO John Bennett noting that passwords persist as a core risk.
The Crisis of Non-Human Identities in Healthcare and Highly Regulated Industries
Highly regulated industries face mounting pressure to secure non-human identities, including service accounts, bots, RPA tools, and AI agents. These identities often outnumber human users 45 to 1, yet 75% lack designated oversight. Misconfigurations, unclear shared responsibility boundaries, and configuration drift continue to drive risk across cloud environments, especially after migration. Center for Internet Security (CIS) experts emphasize that healthcare organizations must strengthen HIPAA compliance and adopt security by default approaches within the shared responsibility model to protect patient data.
Source: Healthcareinfosecurity
