Final call! Black Hat USA, a premier cybersecurity event, kicks off next week, and late registration ends July 31.
A10 + @TrojAISec ...visit us at booth #5548
Become an A10 Agent
Black Hat discount code: A10Networks for up to a $200 discount. https://t.co/42BnHRDD9G
Built-in LLM safety features protect the model.
An AI firewall protects everything around it, inspecting natural-language traffic for prompt injection, jailbreaks, data poisoning, and system prompt leakage in real time.
Watch how AI guardrails work.
▶️ https://t.co/OYR6fpjBqj
The A10 AI Firewall blocks unsafe prompts inline, in real time, while legitimate requests pass through uninterrupted.
This demo walks through exactly how. ⬇️
https://t.co/cTbV3Cm4dz
A watermark invisible to human reviewers automatically triggered a positive review when scanned by a model. The attack required no code exploit, just hidden data in the input.
A10 experts explain why training data integrity is central to AI security. ⬇️
https://t.co/Q92mIE5var
Fake referral accounts, thousands of automated signups, and credits funneled to one master account.
That's business logic fraud at scale, but standard defenses often miss it entirely.
A10's Gary Wang shares two strategies to build a layered defense. https://t.co/7w3HNDFWzQ
The Infosecurity Government Forum provides a valuable platform to discuss the latest cybersecurity challenges, digital resilience strategies, and practical approaches to protecting critical public-sector infrastructure.
We want to thank our summer 2026 interns for their hard work in advancing A10's security and AI technologies. We wish you the best for the coming academic year! 😃
Thank you,
* Kunj Shah
* Abinav Chari
* Candice Chen
* Aaron Kirk
* Sritha Kondragunta
* Jeffrey Guo
* Wilson K.
OWASP now runs two separate AI security frameworks. The LLM Top 10 covers risks like prompt injection and data leakage, while the new Agentic Top 10 covers execution risks once agents hold credentials, memory, and the ability to act. https://t.co/TiN4fnoqEF
Post-quantum cryptography is now in the A10 SSL/TLS data plane. ACOS 7.0.3 adds hybrid KEM support via OpenSSL 3.5: X25519 + ML-KEM-768, secp256r1 + ML-KEM-768, secp384r1 + ML-KEM-1024.
Enabled by default on client-side TLS; no new hardware needed. https://t.co/0dEfeN31Cl
AI gives security teams speed and scale. But human judgment, context, and critical thinking are still irreplaceable.
The strongest security posture combines both.
Read more: https://t.co/ez8DOSoCDz
Defending applications means focusing on who and what is attacking them, and ThreatX is built around that outside-in philosophy.
A10's Carlo Alpuerto explains why inline, real-time protection across on-prem and cloud environments is a valid defense. ▶️ https://t.co/oelhsFX6mq
The second in a series, we held a Korean regional partner seminar with BrainNetworks to introduce A10's business to partners in the Busan, Ulsan, and Gyeongnam regions.
Thank you to BrainNetworks and the 15 participating partners for the insightful discussions and collaboration.
Encrypted traffic captured today can be stored and decrypted by future quantum computers. The time to address that risk is before quantum computing matures, not after.
A10's ACOS 7.0.3 introduces hybrid PQC support, with no new hardware required. https://t.co/PhOAHRu6mv
Adversaries are capturing encrypted traffic today to decrypt it once quantum computing catches up.
ACOS 7.0.3 introduces hybrid KEM support, enabling the parallel execution of classical and post-quantum algorithms; no new hardware required. ➡️ https://t.co/PhOAHRu6mv
DEFSEC, a key event for the Indian defense industry, was a great opportunity to exchange insights with industry experts and explore how advanced cybersecurity technologies can support national and enterprise security initiatives.
Security evolved from firewalls to endpoints to databases, each shift adding new complexity.
AI combines every challenge that came before it into one, and adds entirely new ones on top.
A10's security experts explain this evolution in cybersecurity.
▶️ https://t.co/fKrQTfOYuP
Enabling AI in government requires infrastructure built for the unique security, compliance, and operational demands of the public sector.
A10's Jamison Utter and Redmond Magazine's John K. Waters share the essential steps to get there.
On-demand. ▶️ https://t.co/n9CQTV2HhF
Context-aware AI threat detection uncovers attacker intent, not just attacker actions, and continuously adapts as attack methods shift.
Learn more about how a three-pronged, AI-enhanced security platform excels where traditional firewalls stop short. ➡️ https://t.co/wtTTNxStj4