Large language models are stirring up some phenomena in recent months. So inevitably I was playing ChatGPT too over last weekend, to probe whether it would be a complimentary to some BERT based "traditional" AI chatbots I was knocking up, or rather would it simply sweep them away.
By update