Cycraft co-founders (left to right): Benson Wu (CEO), Jeremy Chiu (CTO) and PK TSUNG (CISO) lead the mission of building the world’s most advanced AI security platform.
Trustworthy AI is important
The transformational power of large-scale language models (LLMS) results in significant security uncertainty, requiring businesses to urgently protect AI models from malicious attacks such as rapid injection, rapid extraction, and jailbreak attempts. Historically, AI security is not a basic feature but an “optional add-on” that exposes valuable AI and data. This surveillance can compromise sensitive data, undermine the stability of the service, and undermine customer trust. Cycraft emphasizes that “AI security must be a standard feature rather than an optional add-on,” and believes that providing stable, reliable, intelligent services is paramount.
The Impending Need for Proactive AI Defense
The need for immediate and effective AI security is more important than ever. As AI becomes more and more embedded in core business operations, the offensive surface expands exponentially, and aggressive defense becomes an absolute necessity. Cycraft leverages its extensive “combat-tested expertise across key domains, including government, finance and tech manufacturing, to accurately address these emerging AI-specific threats. XecGuard’s development means a shift from “treating cybersecurity challenges with AI” to “using AI to protect AI,” ensuring security and resilience are built into day one.
“AI security must be a standard feature, not an optional add-on,” CEO Benson Wu emphasizes Xecguard’s resilience and the integration of experience from key sector defenses. “In the past, we have been able to tackle cybersecurity challenges using AI. We are now using AI to protect our AI,” says CTO and co-founder Jeremy Chiu, who has enabled businesses to confidently adopt AI and provide reliable services. CISO Pk Tsung concluded that “along with Xecguard, businesses can embed security and resilience from day one.”
Cycraft Solution: Xecguard enhances secure AI deployment
Cycraft is leading the global launch of Xecguard, the industry’s first plug-and-play LORA security module dedicated to protecting LLM. Xecguard provides rapid injection, rapid extraction, and robust protection against jailbreak attacks, ensuring enterprise-grade resilience of AI models. Its seamless deployment allows for immediate integration with any LLM without architectural changes, and powerful autonomous defenses out of the box. Xecguard is available as a SaaS, an Openai-compatible LLM firewall on the cloud (such as AWS or CloudFlare Workers AI), or as a built-in firewall for on-premises, or as a custom LLM server with NVIDIA. Strictly validated on major open source models such as the Llama 3B, QWEN3 4B, GEMMA3 4B, and DeepSeek 8B, allowing you to consistently improve security resilience while maintaining core performance, allowing even small models to provide protection comparable to large commercial-grade systems.
Even small models gain corporate-level defense and approach large commercial-grade performance. 
Real-world verification in collaboration with NVIDIA partner APMIC integrates XecGuard into the F1 open source model and shows an average improvement of 17.3% in the overall security defense score for a given attack scenario via LLM Red teaming exercises. With Xecguard and Safety LLM Service, Cycraft offers enterprise-grade AI security, accelerates the adoption of resilient and trustworthy AI across the industry, encouraging organizations to safely deploy AI, protect sensitive data, and drive innovation with confidence.
To learn more about how Xecguard protects LLMS and request a demo, visit www.cycraft.com/en/xecguard.
Hashtag: #cycraft #llmfirewall #aisecurity
https://www.cycraft.com/
https://www.linkedin.com/company/cycraft/
https://x.com/cycraft_corp
The Publisher is solely responsible for the content of this announcement.
About Cycraft Technology
Cycraft It is a leading AI-led cybersecurity company in the Asia-Pacific region. Trusted by hundreds of organizations in the defense, finance and semiconductor industries, AI is designed to prevent, preempt and protect cyber threats. Our expertise is recognized by top institutions such as Gartner and IDC, and is featured at prestigious global conferences such as Black Hat, Defcon, EMNLP, and Code Blue.
