Artificial intelligence (AI) is a series of co-related queries to retrieve data from a series of databases – often times, guessing what the end user needs.

Structure Query Language (SQL) is the most basic form of retrieving data and managing the databases that AI needs.

Although AI could possibly write the query you need, there is a chance that you would need to edit it and/or improve it. After all, some AI models anticipate what the user asks verbally and try to answer right away. The latter is annoying since these AI models use data compiled by user group data – collected from people from a certain age range, geographical location and/or other factor. As such, when a person asks for information that is not in the expected dataset (for example, a question about music sung in Russian when the model expects everyone to listen to music sung in English, worse yet without the compensation for bad pronunciation), AI returns pathetic answers. AI fails to guess. This is where you (as the developer) come in and improve the queries that AI might generate for you.

Many vendors offer their own flavors of AI. For example, MySQL has its own libraries and facilities for AI – Automated Machine Language (AutoML) and Generative AI (GenAI).

MySQL AI provides integrated, automated, and secure machine learning (ML) and generative AI capabilities. AutoML simplifies ML processes, helping you build, train, and explain ML models without data movement or added costs. Similarly, in-database LLMs, a built-in vector store, and embedding models enable GenAI, semantic search, and retrieval-augmented generation (RAG) at reduced infrastructure costs and without data movement.
https://www.mysql.com/products/mysqlai/

As a developer, you need to clean what AI wrongfully believes is correct – often times going back to basic SQL. This is yet another reason why programming is fun.

This entry is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International license.