Novel TTPs

Definition
An AI model to detect articles including attack techniques or procedures that are documented for the first time globally, introducing methods or steps never previously observed in the threat landscape.

Plan

Feedly Threat Intelligence

Includes

new tactics, techniques and procedures

Excludes

new threat actors, new malware, new victims, new campaigns, new defensive techniques, evolutions of known techniques

12x more relevant than simple keyword searches

Researchers have discovered a malicious campaign using a never-before-seen technique for quietly planting fileless malware on target machines.

The second variant deploys Kidkadi, an additional loader implementing a previously undocumented PE injection technique called Vectored Overloading.

The attackers used an innovative technique that manipulated a legitimate Windows tool to run files hosted on a malicious WebDAV server under their control.

Akamai Shares New Techniques for Defenders to Shutdown Cryptominer Attacks

The first and biggest is that we have developed a new method of capturing more evidence of malware sample detection during the test in the form of screenshots taken several times a minute.

Prompt injection is not a new attack technique, but it’s definitely entering a different dimension with the rise of AI agents.

API access
Novel TTPs will appear in an article's commonTopics field when flagged by Feedly AI.

{
  "commonTopics": [
    {
      "id": "nlp/f/topic/novel-ttps",
      "label": "Novel TTPs",
      "salienceLevel": "about",
      "mentions": [
        {
          "text": "Security researchers have unveiled a dangerous new technique that allows malware to completely hide its tracks by faking Windows call stacks a method designed to bypass modern endpoint detection systems."
        }
      ],
      "type": "mitreAttack"
    }
  ]
}

Browse Models