<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI Security on securecode.dev</title><link>https://securecode.dev/categories/ai-security/</link><description>Recent content in AI Security on securecode.dev</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Fri, 01 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://securecode.dev/categories/ai-security/index.xml" rel="self" type="application/rss+xml"/><item><title>AI Security Risks in SaaS Platforms</title><link>https://securecode.dev/insights/ai-security-risks-in-saas-platforms/</link><pubDate>Fri, 01 May 2026 00:00:00 +0000</pubDate><guid>https://securecode.dev/insights/ai-security-risks-in-saas-platforms/</guid><description>&lt;p>AI-enabled SaaS features often get reviewed as isolated prompts or model calls, but the meaningful risk usually sits in the surrounding application workflow. Permissions, tenant boundaries, tool execution, retrieval pipelines, and output handling all matter more than the model alone.&lt;/p></description></item><item><title>Secure AI Integration Patterns</title><link>https://securecode.dev/insights/secure-ai-integration-patterns/</link><pubDate>Tue, 14 Apr 2026 00:00:00 +0000</pubDate><guid>https://securecode.dev/insights/secure-ai-integration-patterns/</guid><description>&lt;p>The safest AI integrations do not rely on the model to be correct, aligned, or cautious. They assume the model can be manipulated, can hallucinate, and can generate plausible but unsafe output. The surrounding system is what turns those limitations into manageable engineering risk.&lt;/p></description></item></channel></rss>