The Modern .NET Show · Hands-On Engineering Podcasts · July 2026
Nixon had spent several minutes on why handing database queries to a non-deterministic model is dangerous — not because it might delete your data, but because it can hand back a confident, wrong answer you then act on. His closing line reframes the risk: the model isn't malicious, it just gets it wrong while you make decisions on the result.
So let's dive into it because this is a deep subject. It really is.
…of grepping and sending HTTP requests and trying to connect to the database yourself or anything like that. Use this tool. The tool knows how to do it. It will help you to achieve that solution. Just to sort of level set, is that kind of, am I on the right, the right tracks as to what an MCP is?
…talking about is a query that gets executed against my data, returned back as a response that looks correct and is presented to me by a model that is disproportionately confident in that data when it's wrong. And now I look at that data and I make decisions in my business thinking that this data that I am using is revealing information inside of my data set when actually it is misleading me. It's not really a question or a category of maliciousness. You know, it's not like the model is trying to ruin your business. The model just got it wrong and now it is ruining your business.
If I can just pause there then and talk about what I'm hearing then. So the SQL MCP Server allows me to have my agent via the LLM interact with the data plane of my data. And say, for instance, if I have an internal tool that I release out to the data people or maybe the people in my business who…
That's right.
That can talk to the database. And I could say in natural language, get me, for instance, the sales by category for the last quarter and create a just get me those, right? Put it into a JSON table so I can display it on screen. So then maybe I can export that to, say, Excel or something like that…