Today we published a comparison between ALMX-1, @almanax's first AI model, Slither, a popular static analysis tool for blockchain code, and human security experts. 

The results are very promising and show that the performance of automated tools for vulnerability detection is much closer to human performance than it used to be just a few months ago. 

Excited for the work that the team has been doing, we have now a good understanding of how to bring these results even closer to human performance. 

Link to the blog available in the comments.

Impressive progress in AI for vulnerability detection! Excited to see ALMX-1's performance compared to human experts. Looking forward to reading the detailed comparison on your blog. Keep up the great work!