And since the data scientist cannot verify the very complex AWK output that should be 100% compatible with his SQL query, he relies on the GPT output for business-critical analysis.

Only if your testing frameworks are inadequate. But I belive you could be missing or mistaken on how code generation successfully integrates into a developer and data scientist's work flow.

Why not take a few days to get familiar with AWK, a skill which will last a lifetime? Like SQL, it really isn't so bad.

It is easier to write complex queries in SQL instead of AWK. I know both AWK and SQL, and I find SQL much easier for complex data analysis, including JOINS, subqueries, window functions, etc. Of course, your mileage may vary, but I think most data scientists will be much more comfortable with SQL.

Many people have noted how when using LLMs for things like this, the person’s ultimate knowledge of the topic is less than it would’ve otherwise been.

This effect then forces the person to be reliant on the LLM for answering all questions, and they’ll be less capable of figuring out more complex issues in the topic.

$20/mth is a siren’s call to introduce such a dependency to critical systems.