> For the complete documentation index, see [llms.txt](https://docs.aquilax.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.aquilax.ai/blog/ai/sast-and-ai-intersection.md).

# SAST and AI Intersection

How we use SLM with SAST

<figure><img src="https://53914109-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FjAmSnvnfbHl4EDK56iDo%2Fuploads%2Fgit-blob-432f408e1380a22775390b73c7ff75a697e5e237%2FSASTvsAI.png?alt=media" alt=""><figcaption></figcaption></figure>

Let’s keep it simple.

Any SAST tool will give you a list of findings. Some will be true issues (True Positives), others won’t (False Positives), and a few might still be under review.

Now imagine scanning the *same* code again—but this time, using a different technique. You still sort the findings into the same buckets: FP, TP, and “needs review.”\\

But here’s where it gets interesting.

What if you combine both techniques? You get a few big benefits:

1. Fewer False Positives – You can catch and eliminate them before they even show up in reports.
2. Access to True Negatives – You see safe areas of the code that one technique alone might have missed.
3. Train the AI to focus only on real issues (TPs) – It starts learning what actually matters.

That’s the idea. And yes, we made it work.