Thoughts

This post has been brought to you by the “How's this for a rant?” department.

Today, I only want to make a small rant concerning artificial intelligence and the vocabulary trends I've spotted in social media and company communications, especially on LinkedIn (Fuck LinkedIn by the way).

What's the problem with using “AI” everywhere? The problem is that everything gets categorized in that umbrella category.

In the 2010s, AI was not very known, and it specifically grouped deep and reinforcement learning, two very well known categories of algorithms.

In the 2020s, ChatGPT was released, and the term AI shifted to now group generative text, then generative images and sound algorithms. All of those are now regrouped under the term “GenAI”.

Until here, everything seems fine. The problem appears in the marketing departments of companies. Since those generative algorithms are new, and also quite spectacular in some cases, we often see “AI-driven” products that don't have anything in common with GenAI. This is the case in my field of work, combinatorial optimization, where AI is used vaguely to mention algorithms that we have known since the 1950s!

So, this post is not adressed to the general public, but to the marketing departments that ride on the vague usual definition of AI: stop misleading people with AI-driven products simply because AI makes your spitch shiny. Use the correct terms where they apply.

---

Thank you for reading. Reach me for further discussion and updates!