BBVA Labs recently tested OpenAI’s GPT-3, one of the largest Artificial Intelligence (AI) language models ever created, to generate summaries of bank customer conversations in Spanish. Without any prior customization or re-training, they asked it to create useful summaries for agents joining or rejoining a conversation with a customer. This could save time and improve the customer experience. The tests were done on ten unedited Spanish conversations about banking taken from online forums. The results showed that GPT-3 was capable of summarizing complex conversational threads accurately and quickly enough for human agents to use them in their workflows.