Expertise

We fine-tuned Llama3-8B to improve its resistance to indirect prompt injection attacks on question/answer tasks involving emails and articles. Reproducible fine-tuning scripts, along with the model and a quantized version, are available on Hugging Face and Ollama for testing and experimentation.

This comprehensive analysis uncovers the business mechanics of ransomware groups, their ecosystems, and how they interact. The report not only provides a clear picture of the current ransomware situation but also reveals some unexpected trends.

In our latest research paper, Tim West, W/Intelligence’s Director of Threat Intelligence and Outreach, has looked into the cyber threats facing the Paris 2024 Olympics. As one of the world’s greatest events, there is always going to be unwanted interest, but how much should we worry and who are the main antagonists?

This report will explore the trend of mass exploitation of Edge Services and Infrastructure and will put forward several theories as to why they have been so heavily and successfully targeted by attackers.

We demonstrate how Google’s Gemini Advanced can be coerced into performing a social engineering attack. By sending a malicious email, attackers can manipulate Gemini's responses when analyzing the user's mailbox, causing it to display convincing messages that trick the user into revealing confidential information from other emails.