Beyond simply a “chat,” we’ve embedded AI across the solution - from human-readable, automated analysis, to instant generation of high-quality intelligence reports and an AI assistant that follows your work, providing key insights in whichever context or activity you’re in. While other generative AI solutions rely on simple integrations with ChatGPT, Cybersixgill IQ leverages AI across its capabilities, enhancing every step of the intelligence process.

Our generative AI tool transforms how organizations can leverage threat intelligence in three key ways: Intel analysis - Turning tactical, raw intelligence, previously only accessible to experienced analysts, into more easily understood, context-and-insight-rich summaries that are usable and actionable for every cybersecurity professional. Intel generation - Using AI to automatically and dynamically generate refined, finished intelligence customized to each organization’s industry, geography, user persona, and business needs. Intel experience – Delivering insights via a threat intelligence chatbot, a context-specific assistant, or across the Cybersixgill portal.

Off-the-shelf and open-source AI solutions are only as good as the data they have access to, and they don’t have access to our sources. If you ask ChatGPT a question about activities on a deep web forum or a dark web market, you won’t get a response because it doesn't have access to these sources. Access to unique, underground data is one of the differentiating aspects of Cybersixgill IQ. Our data is continually updated, whereas ChatGPT’s data only goes through September 2021 and we include information about IOCs, which ChatGPT doesn’t do.

We continue to implement our solutions with a security-first approach. For example, when performing generative AI processes that involve external services such as ChatGPT, we protect the confidentiality of our customers by masking sensitive data and local processing, ensuring that our customer asset information is never shared with OpenAI/GPT/LLM.

To enable AI-driven chat with organizational context while ensuring customer data safety, we have developed a mechanism that separates the prompt sent to OpenAI from the internal analysis of the organization's context. First, IQ queries OpenAI for a general answer to the question (e.g., "What are the MoveIT CVEs?") without including customer context (assets). Then, we run an internal Cybersixgill service on our servers to analyze and match the AI response with your assets (e.g. whether you have any of the MoveIT CVEs in your assets). All of this occurs within our local environment, maintaining the highest security standards.

When off-the-shelf LLMs like GPT and Bard lack access to sources that provide the right information, they may occasionally generate fake or “hallucinated” content. Cybersixgill IQ does not do this. Our solution has been scoped and fine-tuned to prevent it from making up answers. We also train our proprietary transformer (GPT) CTI model to fit our specific domain, unique data, and customers' use cases. By taking this extra step, we can leverage the benefits of large language models and transformer technology to fit the distinct context of threat intelligence.

Cybersixgill’s expert intelligence professionals and data analysts continuously review the model and its results, performing qualitative and quantitative analysis to continuously improve the model. Additionally, we regularly request customer feedback and respond accordingly within a few days. 

Yes. With Cybersixgill IQ, you can ask questions and receive answers in multiple languages. For example, if you ask IQ about emotet (malware) in German, your response will be generated in German.

Cybersixgill IQ is embedded into our products and enhances the overall value of our CTI solutions. The chat, item summarization and threat entity descriptions are included as part of the portal with no additional cost.