The State of AI Security | Cisco Report

Artificial intelligence (AI) has emerged as one of the defining technologies of the 21st century. It has transformed both our personal and
professional lives, and its rapid advancement will continue to reshape the ways in which
businesses operate. 

The State of Industrial AI | Cisco Report 

AI and browser threats stand out in the 2026 Threat Detection Report

Red Canary's annual analysis brings a year’s worth of security operations and intelligence insights, including how adversaries are both leveraging and targeting AI.

Mythos-Class Cyber: Tales from the Trenches

Why data centres are reshaping resilience and cybersecurity for critical infrastructure

"Protecting data centre infrastructure and coordinating incident response with grid operators in advance can support grid stability during cyber events or power disruptions."

The AI infrastructure reckoning: Optimizing compute strategy in the age of inference economics

The cloud-first era is giving way to a new hybrid reality as enterprises race to align their infrastructure with the true demands of production-scale AI.

Cyber risk in 2026: What executives must know about AI, fraud, geopolitics and more

Drawing on the WEF's Global Cybersecurity Outlook 2026, this article breaks down the three forces executives must understand heading into 2026: AI's expanding role in cyber risk, geopolitical fragmentation, and cyber-enabled fraud now surpassing ransomware as the top CEO concern.

The focus is shifting from excitement about AI agents to understanding how agentic AI technologies are maturing.

"The need for oversight and discipline is becoming evident early in the adoption cycle — not only after large‑scale deployment."

Deterministic Code-in-the-Loop

Human approval workflows weren't built for machine-speed attacks. Learn how to replace them with deterministic code that makes the right call at every decision point without waiting on a human in the loop.

The Death of Security by Obscurity

Staying small stopped being a defense strategy. Understand what the end of obscurity-based security means for your organization and what a modern posture looks like in its place.

AI-Scale Attacks: The New Normal

Mythos-class AI has put world-class offensive capability within reach of any attacker. Learn what that shift means for your threat model and how practitioners are adapting their defenses.

AI Drift & the Claude Code Wake-Up Call

AI drift doesn't announce itself. Learn what it looks like when a model quietly diverges from your intent and how to build the checkpoints that catch it before it becomes a real problem.

Patch Exceptions as Attack Surface

Unresolved patch exceptions were always a risk. AI turned them into exploit chains. Learn how attackers are connecting the dots across your exception policy and how to close the gap.

Agentic AI & Prompt Injection Risk

Prompt injection in agentic environments doesn't always look like an attack. Learn how adversarial inputs move through deployed agents and what it takes to protect them in production.

State of Agentic AI Security and Governance 2.01

Security at Cisco Live: Going Shields Up For the Agentic Era

AI = GPUs -- The most damaging misconception in enterprise AI 

William Fowler (Intel) breaks down the #1 siloed mindset stalling enterprise AI infrastructure: equating AI with GPUs, and why traditional benchmarks fall apart in agentic workflows. Richard Piasentin then raises the SLA challenge nobody's fully solved. 

GPUs sitting idle -- the agentic paradox nobody talks about

You bought the GPUs. Now they're sitting idle while token costs spike. Intel's William Fowler and Cisco's Richard Piasentin unpack why agentic workflows break traditional GPU utilization, and how to think about the system as a whole instead. 

Agentic AI hits your infrastructure at 100-1,000x human speed

From one-shot LLMs to agentic AI: the infrastructure evolution explained

GPU, CPU, TPU, ASIC -- what's the difference and why does it matter now?

The token shock cycle: we're repeating the cloud migration mistake with AI

Why the AI chip you deploy today may run on a CPU tomorrow 

SLMs on CPUs today: what's actually running and how small is small enough? 

DualPath: Breaking the Storage Bandwidth Bottleneck

Agentic AI Workload Characteristics

Agentic AI Requires More CPUs

AI Agents Can Access More Than You Intended.

There is More AI in Your Company Than you Think. 

AI is Exposing the Gaps in Your Security Program.