#response-selection

#response-selection

This operation is about a set of very specific objectives; the president laid them out on the very first night of operations. I'll repeat them to you now because I hear a lot of talk about we don't know what the clear objectives are.

The majority of AI products remain tethered to a single, monolithic UI pattern: the chat box. While conversational interfaces are effective for exploration and managing ambiguity, they frequently become suboptimal when applied to structured professional workflows. To move beyond "bolted-on" chat, product teams must shift from asking where AI can be added to identifying the specific user intent and the interface best suited to deliver it.

By comparing how AI models and humans map these words to numerical percentages, we uncovered significant gaps between humans and large language models. While the models do tend to agree with humans on extremes like 'impossible,' they diverge sharply on hedge words like 'maybe.' For example, a model might use the word 'likely' to represent an 80% probability, while a human reader assumes it means closer to 65%.

Something I've been noticing a lot lately is that the confidence of AI chatbots is getting in the way of the communication between human and machine. Chatbots spit out false information with such confidence that it conveys the idea that the information is true, even though the chatbot has little to no evidence for it - yet that fact is never communicated.

Google Search's AI makeover continues. The company said that, starting today, mobile users will be able to ask follow-up questions to AI Overviews, Google's AI-generated search summaries. Doing so will launch users into a back-and-forth with AI Mode, its more conversational take on search that already lives in a separate tab on the search page. After Google's AI Overviews awkwardly stumbled out the gate in 2024 ( pizza glue, anyone?) they've gradually become a staple of the Search experience.

The new talk of the town is one where humans have no place a site called Moltbook that describes itself as a "social network for AI agents." The Reddit-styled site, launched in late January by US-based entrepreneur Matt Schlicht, is one where thousands of AI assistants talk to each other and discuss topics ranging from the technical to the philosophical.