Blog posts by Jessica Brennan
Senior Product Marketing Manager, AI Security | F5
Blog posts written by Jessica Brennan(33)
From one app to a fleet of agents: A 90-day AI security framework
Skip the costly “deploy now, secure later” approach with a comprehensive framework for protecting AI models, applications, and autonomous agents.
By James White, Jessica Brennan
AI latency: Why security can’t be a trade-off
AI latency isn’t just about speed. It’s about designing systems that balance performance, security, and reliability.
By Jessica Brennan
대화는 방화벽으로 막을 수 없다: AI 레드팀 활동이 핵심 과제로 떠오른 이유
AI 배포가 기계적인 속도로 빠르게 진행되면서, 기존의 소프트웨어 테스트 방식은 더 이상 제 역할을 하지 못하고 있습니다.
By Jessica Brennan, Allan HealyYou can’t firewall a conversation: How AI red-teaming became mission-critical
With AI deployments moving at machine speed, traditional methods of testing software are no longer fit for purpose.
By Jessica Brennan, Allan Healy
Securing enterprise AI in financial services
Learn why financial institutions adopting AI must also implement guardrails that can be customized and refined over time.
By Jessica Brennan
F5 AI Remediate: Closing the AI security gap
F5’s new offering, F5 AI Remediate, shortens the path from AI vulnerability discovery to deployable protection.
By Jessica Brennan
AI observability: Auditing and tracing AI decisions
Discover how AI observability ensures accountability for AI decisions with advanced runtime visibility, auditing, & traceability across AI workflows & systems.
By Jessica Brennan
AI compliance and regulation: Using F5 AI Guardrails to meet legal and industry standards
See how F5 AI Guardrails and F5 AI Red Team simplify compliance, reduce legal exposure, and enable organizations to demonstrate responsible AI governance.
By Jessica Brennan
Classifier-based vs. LLM-driven guardrails: What actually works at AI runtime
Not all AI guardrails work the same way. Learn why it’s important to know the distinction between classifier-based and LLM-driven guardrails.
By Jessica Brennan
AI security through the analyst lens: insights from Gartner®, Forrester, and KuppingerCole
Enterprises are discovering that securing AI requires purpose-built solutions.
By Jessica Brennan
F5 recognized as one of the Emerging Visionaries in the Emerging Market Quadrant of the 2025 Gartner® Innovation Guide for Generative AI Engineering
We’re excited to share that F5 has been recognized in 2025 Gartner Emerging Market Quadrant(eMQ) for Generative AI Engineering.
By Jessica Brennan
Closing the Loop: Why AI Security Remediation Matters
AI security remediation is one of the most important disciplines in enterprise cybersecurity. As organizations adopt GenAI at scale, they face a flood of new vulnerabilities. But without effective rem...
By Jessica BrennanSmarter Guardrails, Stronger Security with the New AI Assistant
F5's new AI Assistant is built directly into AI Guardrail's custom scanners feature. This feature allow enterprises to go beyond out-of-the-box protections by leveraging GenAI to create policies that ...
By Jessica BrennanModel Guardrails Aren’t Enough: Why Runtime Guardrails Are Essential for AI Security
Model guardrails alone can’t keep AI systems safe. This is something our research team sees every month when testing frontier models (results published on our AI Security Leaderboards). While model gu...
By Jessica BrennanHow Forrester’s AEGIS Framework Validates an Inference-First Approach
The rise of agentic AI is transforming the enterprise security landscape. These self-directed AI agents operate autonomously, make real-time decisions, and can interact across systems in ways traditio...
By Jessica Brennan커스텀 스캐너 버전 관리: 최적의 민첩성을 위한 AI 방어 통제의 진화
AI 애플리케이션과 에이전트를 배포하는 기업은 진화하는 사용 사례, 정책 및 규정에 맞춰 보안을 유지할 수 있는 민첩성이 필요합니다. F5 AI Guardrails를 통해 조직은 이미 위협을 차단하고 데이터 유출을 방지하며 보안 정책을 실시간으로 적용하는 강력한 방어용 Guardrails를 활용하고 있습니다. 이제 Custom Scanner Versioni...
By Jessica BrennanCustom Scanner Versioning: Evolving AI Defensive Controls for Optimal Agility
Enterprises deploying AI applications and agents need agility to keep security aligned with evolving use cases, policies, and regulations. With F5 AI Guardrails, organizations already gain powerful de...
By Jessica BrennanAgentic Signature Attack Packs: Red Team AI for the Next Generation of Security
Every new capability that GenAI introduces comes with new vulnerabilities. As a result, companies must red-team AI systems regularly to proactively uncover weaknesses and secure AI applications before...
By Jessica BrennanVisualize the Attack: Gen AI Pentesting Gets an Upgrade with Agentic Fingerprints
You can’t defend what you can’t see. That’s the problem with Gen AI pentesting today. It tells you whether an attack worked, but not how it unfolded or why your defenses failed. For security leaders t...
By Jessica Brennan결과 분석과 이를 통한 AI 위협 탐지 개선
CalypsoAI의 AI 보안 솔루션 Inference Defend에 AI 위협 탐지 워크플로의 가시성과 제어력을 강화하도록 특별히 설계된 새로운 기능인 Outcome Analysis가 추가되었습니다. Defend의 실시간 보호 계층에 포함된 Outcome Analysis는 보안 팀이 스캐너(또는 가드레일)가 프롬프트나 응답을 플래그하거나 차단한 정확한 이...
By Jessica BrennanOutcome Analysis and How It Improves AI Threat Detection
CalypsoAI’s Inference Defend AI security solution has a new feature, Outcome Analysis, that is purpose-built to enhance visibility and control in AI threat detection workflows. As part of Defend’s rea...
By Jessica BrennanWhat the Forrester Budget Planning Guides Reveal About the Future of GenAI Security in 2026
As enterprises look ahead to 2026, two new Forrester Budget Planning Guides: Budget Planning Guide 2026: Security And Risk and 2026 Budget Planning: Prepare For Even More Volatility, offer valuable in...
By Jessica BrennanLLM evaluation: Building trust with security scoring
As enterprise AI adoption accelerates, LLM evaluation is no longer just about performance, it’s about trust. Security teams must validate not only how well a model performs, but how safely it behaves....
By Jessica BrennanLLM 평가: 보안 점수화를 통한 신뢰 구축
기업의 AI 도입이 가속화되면서 LLM 평가는 더 이상 성능만의 문제가 아니라 신뢰의 문제가 되었습니다. 보안팀은 모델의 성능뿐만 아니라 얼마나 안전하게 작동하는지도 검증해야 합니다. 하지만 대부분의 평가 프레임워크는 정확도 벤치마크에 초점을 맞추고 있어 보안 및 위험 요소는 모호하거나 측정되지 않은 채로 남아 있습니다.
By Jessica BrennanPrompt injection attacks: What you need to know
Prompt injection is the top AI security threat. Learn how it works, why it’s dangerous, and how enterprises can secure AI systems against it.
By Jessica BrennanEnterprise AI Security: Key Insights from Forrester’s Recent Report
Why AI Security Posture Management Is Now a Top Priority According to Forrester’s recent research Key Trends In AI Detection Surface, organizations must urgently rethink how they manage and secure AI ...
By Jessica BrennanWhat Your AI Red Team Is Missing And How to Fix It
Red-teaming has long been a cornerstone of enterprise security but when it comes to generative AI, most organizations are applying outdated tactics to a new kind of threat landscape. Static prompt tes...
By Jessica BrennanHow to Build a GenAI Policy That Actually Protects Your Business
GenAI usage is exploding across teams and tools, but most enterprises are still missing the controls to secure it. Security isn’t failing because there is a lack of policies. It’s failing because orga...
By Jessica BrennanSelf-Hosting vs. Models-as-a-Service: The Runtime Security Tradeoff
As GenAI systems continue to move from experimental pilots to enterprise-wide deployments, one architectural choice carries significant weight: how will your organization deploy runtime-based capabili...
By Jessica Brennan보안을 강화하고, 확장하고, 성공하라: 런타임이 최우선인 이유
AI는 더 이상 실험 단계에 머물러 있지 않습니다. 이미 실제 운영 환경에서 챗봇을 구동하고, 의사결정을 지원하며, 기업의 운영 방식을 재편하고 있습니다. 하지만 눈앞에 드러나 있는데도 간과되는 영역이 있습니다. 바로 런타임입니다. 학습이 주목받는 동안, AI가 실제로 작동하는 곳은 런타임입니다. 모델이 실제 사용자, 데이터, 의사결정과 상호작용하는 곳이기...
By Jessica Brennan