Almost half of all code generated by AI which turned out to contain security errors – even the affected large LLMs


  • The report finds 45% of AI-Generated Code had security error
  • Java is the worst offender, python, c# and javascript touched too
  • Rise in vibe -coding could make these threats even worse

Almost half (45%) of AI-generated code contains security errors despite the fact that they appear that are ready, new research from Veracode has found.

Its study of more than 100 large language models across 80 different coding tasks revealed no improvement in security across recent or larger models – an alarming reality for companies that depend on AI tools to back up or even replace human productivity.

Leave a Comment

Your email address will not be published. Required fields are marked *

Scroll to Top