This is factually wrong. Where did you get this information from?
We’re using a transformer-based deep learning model (so a deep neural network).
The core model is a transformer‑based sequence model that captures temporal relationships in activities. It has multiple input heads for various types of input data.
There’s a lot of misinformation about this launch and I’m not sure where it’s coming from. Some people think we are using LLMs/ChatGPT and producing “AI Slop”, which is a term specifically tied to LLMS (which we aren’t using).
Some people think we are lying and not doing anything with AI.
Some people think that unless every aspect of their training is managed by AI, none of it is.
And some people think that Machine Learning isn’t AI because it’s not an LLM. 
