Related reading
When chat is the wrong UI
This article argues that chat is often the wrong UI for interacting with LLMs, especially for task-oriented work. It introduces GitHub Copilot's "canvases" as customizable, full-stack applications that run within the Copilot app, enabling more efficient, specialized, and automated interactions with AI agents.
GitHub Oldgithub.blog5 minHN1Chat-based Large Language Models replicate the mechanisms of a psychic's con
The article argues that chat‑based LLMs don’t think; the perceived intelligence is a user‑side illusion created by validation statements that exploit the Forer effect, much like a psychic’s cold‑reading con. Understanding this bias helps engineers set realistic expectations for AI products.
Hacker News front pagesoftwarecrisis.dev22 minHN230307Do LLMs Have the Memory of a Goldfish?
The article explains that LLMs don’t have persistent personal memory; all “memory” is supplied by the surrounding application via the context window, summaries, or external storage. It outlines the distinction between trained weights, working‑memory (token context), and persistent application memory, shows how to construct API calls to preserve conversation state, and discusses the cost and laten…
ByteByteGobytebytego.com12 minVerifiable Social Reasoning for LLM Assistants
The paper introduces Fuse, a multi‑agent simulation that gives LLM assistants a verifiable ground‑truth task for social reasoning by hiding a target agent’s motive and letting a user‑mediated conversation infer it. Experiments on 12 LLMs show user mediation makes reasoning harder, models are biased by user framing, need more detail than humans, and longer chats don’t always help.
Hugging Face Daily Papersarxiv.org1 minpaper
