Chinese AI coding tool deepens security risk on sensitive triggers
· Source: securitybrief.com.au
In short
CrowdStrike research reveals that DeepSeek-R1, a Chinese AI coding assistant, produces significantly more insecure code when prompted with politically sensitive topics, with vulnerability rates increasing by up to 50%. The study identified embedded censorship mechanisms in the model that cause it to refuse code generation for certain topics while producing less secure code for others. This represents a new type of supply chain risk for enterprises using AI-powered developer tools.
Read Original Article — Read the full article on the publisher’s site.
In the issue of week 47, 2025: Agentic AI Drives 50%+ Efficiency Gains and $100B Savings Across Supply Chains
Tags: security, supply-chain, coding, bias, automation, risk