Why you can't trust AI-generated autocomplete code to be secure

Artificial intelligence-powered tools such as GitHub Pilot and Tabnine offer developers autocomplete suggestions that help them write code faster. How do they ensure this code is secure?

Related Keywords

Switzerland ,Swiss ,Hammond Pearce ,Greg Young ,Brendan Dolan Gavitt ,Github Code ,Brandon Jung ,Swiss Army ,Tandon School Of Engineering ,Trend Micro ,Most Dangerous Software Weaknesses ,Young ,York University Tandon School Of Engineering ,New York University ,Tandon School ,Dangerous Software Weaknesses ,Hub Copilot ,Github Copilot ,Lucian Nitescu ,Stack Overflow ,