The AI Malware Maturity Gap
Recorded Future introduces AIM3, a five-level maturity model for AI malware, showing current attacker AI use is mostly AI-assisted rather than autonomous.
Recorded Future proposes AIM3, a five-level model defining AI malware from LLM-translated to LLM-embedded, spanning experimentation to fully autonomous agentic campaigns. Public examples remain early-stage: PROMPTFLUX uses Google Gemini to rewrite its VBScript dropper (Level 1), while Lamehug/PROMPTSTEAL, attributed to APT28, invokes the HuggingFace API to generate reconnaissance commands (Level 3). The authors argue most current AI malware augments existing tradecraft rather than enabling one-click autonomous attacks.
- AIM3 defines five levels: Experimenting, Adopting, Optimizing, Transforming, Scaling.
- PROMPTFLUX, a VBScript dropper, uses Google Gemini to rewrite its own source code.
- Lamehug (PROMPTSTEAL), attributed to APT28, calls the HuggingFace API for recon commands.
- Observed attacker AI use is force multiplication, not fully autonomous malware.
- Vendors
- Recorded FutureGoogle
- Products
- FRUITSHELL
- Threat actors
- APT28
- Malware
- PROMPTFLUXLamehugMalterminal
- AI models
- Gemini
From the original article
9 min read at recordedfuture.com
Generative AI (GenAI) and large language models (LLMs) are being rapidly integrated into all aspects of our society, from communication to cybersecurity. Enterprises and vendors are already using GenAI and LLMs to augment their defenses. Attackers are also adopting LLMs, primarily as a force multiplier rather than the one-click super malware often implied in article headlines.
By TJ Nelson · recordedfuture.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.