# [MED] More accurate than GPT-4: How Snyk’s CodeReduce improved the performance of other LLMs

**Source:** Snyk
**Published:** 2024-05-07
**Article:** https://snyk.io/blog/deepcode-ai-vulnerability-autofixing/

## Threat Profile

Snyk Blog In this article
Written by Eric Fernandez 
May 7, 2024
0 mins read Snyk has been a pioneer in AI-powered cybersecurity since the launch of Snyk Code in 2021, with the DeepCode AI engine bringing unmatched accuracy and speed to identifying security issues in the SAST space for the first time. Over the last 3 years, we have seen the rise of AI and LLMs, which Snyk has been at the forefront of with the introduction of new AI-based capabilities, such as Snyk Agent Fix, our vulnerability au…

## Indicators of Compromise (high-fidelity only)

- _No high-fidelity IOCs in the RSS summary._ If the source publishes a technical write-up with defanged IOCs in the body, those would be picked up automatically on the next pipeline run.

## MITRE ATT&CK Techniques

- _Narrative-keyword inference returned no technique mappings; review article for ATT&CK relevance manually._

## Kill chain phases observed

_(none detected from narrative keywords)_

## Recommended hunts

_No actionable hunts can be derived from the RSS summary alone. The article may still warrant manual review — open the source link for actor attribution, IOCs in the body, and TTP detail._


## Why this matters

Severity classified as **MED** based on: 0 use case(s) fired, 0 technique(s) inferred. Read the full article for actor attribution, tooling details, and any defanged IOCs in the body that aren't visible in the RSS summary.