With the recent concerns being expressed by a multitude of experts in the field of large language models, wouldn't a report detailing FastGPT and Universal Summarizer's inner-workings be appropriate? Especially in a world with hazardous intellectual property laws and the current uncertain nature of this landscape, there could be an opportunity to become a gold standard for people to look to if they worry about what goes on just behind the veil. If people are curious, they can look into what makes Kagi's LLM give the answers it does in the way that it does. Perhaps have a way for users to query the model's database and potentially allow them to opt out of having their website indexed, writing, or other media indexed.
