AI labs are universally terrible at naming things. For example, OpenAI has three different Codex services that do entirely different things. Anthropic has two Max plans. The current best and worst models on ChatGPT are 4o-mini and o4-mini. Xai is named Xai. 
On the other hand, journalists are great at naming things that don’t need to be named. Examples: Harvard Derangement Syndrome, Mommunes, Quiet Qutting. I doubt many people are going to use “Mommunes” to refer to single mothers living together, but it is a great name regardless.

Fortunately, as I cover the local school district for my job, which does not have many stories during the summer, I can offer my services to an AI lab willing to hire me. Although I have wanted to work at OpenAI at age 9 as a software engineer, I have been set on a bad path eventually leading to becoming a New York Times writer and finding names for people who don’t want to be named, but I can use these human journalist skills for good by naming software. 

My other part-time job, AI art advocate on Reddit, has been stolen by soulless slop AI Reddit posters.^[1] Even more sadly, recent steel tariffs have dramatically raised the price of paperclip derivatives, and uncertainty in the chip market has made it inefficient to simulate mussels eating clips, previously the most effective way to raise the global number of perceived paperclips per neuron.

If any hiring manager for an AI lab is reading this, please hire me so I can reset my paperclip hedonic treadmill to pre-tariff levels, and so you can have your software make sense to customers. You can contact me at hdrussell@outlook.com.

To show how good of a candidate I am, here is a list of suggested names to finish off the post.

1. ^{^}

   I don't use LLMs to write articles but I do use it to help me debug code, not for my job.