Cheque fraud detection is big business in the few English speaking countries that still use them for payment. Object detection algorithms with deep learning can transcribe the text from cheques, but humans still write the explanations for a cheque has been marked as being suspected fraudulent.

In this LLM Makerspace edition, we will build a LLM that not only detects whether a scanned image of a cheque is a fraud, but also writes an explanation for why the cheque has been marked as fraud. We will use a fine-tuned open-source LLM, so the entire AI system can run entirely in your data center - it will be a data sovereign LLM application.