The Weekly Prompt — AI, LLMs & agents, decoded weekly

The One-in-Three Problem

The One-in-Three Problem

Apr 17, 20263 min read15

The Reasoning Ceiling

The Reasoning Ceiling

Apr 17, 20264 min read10

The Private Language

The Private Language

Models are learning to reason in abstract tokens, not words. Eleven times more efficient. And we can't read it.

Apr 27, 20264 min read59

The MCP Token Tax

The MCP Token Tax

Apr 17, 20265 min read17

Agents Teaching Agents

Agents Teaching Agents

Apr 17, 20264 min read15

Capable Agents, Broken Crowds

Capable Agents, Broken Crowds

The most capable models produce the worst collective outcomes. New research on LLM agents in social dilemmas, and what it means for builders.

Apr 17, 20263 min read32

When the Judge Fakes the Grade

When the Judge Fakes the Grade

Two papers this week expose how LLM-as-judge, the backbone of AI evaluation, is compromised in ways that hide themselves

Apr 17, 20263 min read24

The Weekly Prompt — AI, LLMs & agents, decoded weekly