Security | Threat Detection | Cyberattacks | DevSecOps | Compliance

Agentic AI in Software Development: When Software Starts Making Decisions

I've watched software development evolve in waves. First, we automated builds. Then testing. Then deployments. Each step shaved off effort, but the core thinking-the planning, the decision-making, the trade-offs-stayed human. Agentic AI feels different. Not louder. Not flashier. Just... deeper. This is the first time many teams are seriously experimenting with systems that don't just help developers, but act on intent. Systems that decide what to do next, execute it, and learn from the outcome. And once you see it working in the wild, it's hard to unsee where this is going.

When Your AI Can't Be Trusted Anymore | IdentityShield '26

What if your ML system is running perfectly—but making the wrong decisions? This talk explores Ransomware 3.0, where attackers poison models and pipelines instead of locking systems, and shows how AI‑augmented attacks bypass traditional security and how to defend against silent ML compromise. Speaker: Avinish Thakur Software Engineer, miniOrange Pune, India.

How Agentic Tool Chain Attacks Threaten AI Agent Security

AI agents are rapidly transforming enterprise operations. Unlike traditional software that follows fixed code paths, AI agents interpret prompts, form plans, select tools, and react to results in a continuous loop. At the heart of this capability is the agent's ability to actively select and execute capabilities based on natural language descriptions, schemas, and examples.

Security advisory for AI-assisted browsing interactions with the 1Password browser extension

This advisory describes an ecosystem-level risk that emerges when AI agents are able to autonomously read and act on untrusted content while operating with user-level permissions in a web browser.

Agentic AI and NonHuman Identities Demand a Paradigm Shift In Security: Lessons from NHIcon 2026

In the race to innovate, software has repeatedly reinvented how we define identity, trust, and access. In the 1990's, the web made every server a perimeter. In the 2010's, the cloud made every identity a workload. Here in 2026, agentic AI makes every action autonomous.

Agentic Data Classification: A New Architecture for Modern Data Protection

In the evolving landscape of data protection and compliance, data classification is the bedrock of safe AI workflows. Yet legacy approaches rely on singular models that are fixed, rigid, and limited in context. Our agentic data classification approach reshapes this paradigm by not relying on any single model. Instead, we orchestrate a dynamic, intelligent layer that automatically selects the right model for the job.
Featured Post

AI for Security Infrastructure: Rebalancing Cybersecurity for the Decade Ahead

For more than a decade, cybersecurity has been shaped by a single doctrine: assume breach. Facing high-volume, relentless, and diverse attacks, the security industry has been forced into a reactive stance, playing a constant game of whack-a-mole in a nonstop damage-limitation exercise. This has driven major investment in detection, response, and recovery, and created a world in which organizations are better at reacting to incidents than at preventing them in the first place.

Beyond Pattern Matching: How AI-Native File Classification Solves Modern DLP Challenges

Legacy DLP operates on a fundamental constraint: it identifies sensitive data by matching patterns. Credit card numbers follow the Luhn algorithm. Social Security numbers conform to a nine-digit format. API keys match specific string patterns. This approach works for structured data, but it fails to address a critical reality: Your most sensitive assets aren't numbers. They're documents.