He’s the problem with using LLMs that take data and analyse it straight out of the box.

← Back to Blog

He’s the problem with using LLMs that take data and analyse it straight out of the box. Sometimes you require an understanding of context and have the experience, general, and tacit knowledge to interpret data to reach a reasonable, useful, and value-added conclusion.

It was a failure of assumption…. that an LLM, no matter how powerful, could understand the data it was working with, simply by slapping a model on top of it.

Without a meaning layer surrounding it all… the definitions, relationships, and business rules that reflect how your organization actually operates, the model is reconstructing your business from what it’s learned across the Internet.

It WILL give you an answer. Every time. Confidently.

And it will be wrong.

Here’s why using AI in cybersecurity and compliance is hard………….

@Thaddeus Dziekanowski Darlene Newman Brian D. McCarthy

@Parava Security Solutions

#cybersecurity #ai

View Original Post
https://www.linkedin.com/feed/update/urn%3Ali%3Ashare%3A7441849642453078018