A forensic tool for backdoored code completions in AI assistants

Researchers have developed a forensic tool to detect backdoored code completions in AI coding assistants, addressing the risk of poisoned training data that can cause models to generate insecure code when triggered by specific cues. The tool helps trace and identify tampered examples that introduce hidden vulnerabilities into AI-generated code.

Developers lean on AI coding assistants for a growing share of their daily work, letting the tools predict the next few lines and accepting many suggestions with a quick glance. Those tools learn from large collections of code, and some of that code can be tampered with before training starts. A poisoned example teaches a model to write insecure code when it sees a certain cue, and the flaw sits quietly until the right prompt … More https://www.helpnetsecurity.com/2026/07/20/tracing-backdoored-code-completions/ The post A forensic tool for backdoored code completions in AI assistants https://www.helpnetsecurity.com/2026/07/20/tracing-backdoored-code-completions/ appeared first on Help Net Security https://www.helpnetsecurity.com .