A map of common AI tasks.

Late 2022, with examples.

Keno Leon
9 min readDec 19, 2022


It’s hard to keep pace with AI these days as new papers, implementations and even consumer facing products seem to sprout periodically, the purpose of this post is to investigate what the main tasks in general are behind current AIs in hopes to get a better sense of the general direction and accomplishments of AIs.

— — — — ⭐️ Support content writers : SUBSCRIBE TO MEDIUM ! ⭐️ — — — —

⚠️ This post should answer the question What’s AI good for today? in simple terms for stakeholder and newcomers alike.


On the “basic” end, AIs are tasked with figuring out relationships between data, it’s implied at this stage that you have some data to train with and once the AI is trained you can use it to predict on new unseen data or get some insight from your analysis.

On the left you have some data points and the AI infers the relationship (red line/right) and predicts new pairs (dashed points/right).

Simple linear regression ( like the one pictured above ) where just a pair of data points fit neatly in a tight line is just the start…