Signals
Production-oriented research signals and interpretation for AI systems builders.
Guides and explainers
Detailed guides and practical technical analysis.
Latest analysis
Recent research, benchmark reviews and technical updates.
Practical tools
Templates for budgets and project planning
BoredTools offers practical spreadsheets for budgets, freelance work and small projects.
Agent Leaderboards Can Be Cheaper Without Being Safer
A March 2026 paper on efficient agent benchmarking found that mid-difficulty task subsets can remove large parts of an agent benchmark while preserving...
Multi-Agent Systems Need Specs Before More Agents
Multi-agent systems are getting easier to assemble and harder to trust. A new June 2026 paper from Cisco researchers argues that the missing layer is not...
Power Grid Agents Need Constraint Tests, Not Chat Scores
A June 2026 power-systems benchmark argues that language-model agents can solve grid-engineering tasks, but the useful signal is narrower: the agent must...
Agent Tool Menus Are a Safety Surface
New agent benchmarks suggest the visible tool menu is not a neutral implementation detail. It changes success, cost, wrong-tool calls, and risk exposure.
Agent Memory Fails on Relationships, Not Recall
New June 2026 memory benchmarks show why long-running agents fail when facts conflict, evolve, or depend on hidden relationships.
SMAC-Talk Shows Agent Chat Is Not Coordination
SMAC-Talk adds natural-language communication and deception to StarCraft-style multi-agent evaluation. The result is a useful warning: agent chat can expose coordination failure as easily as it fixes
Agent Benchmarking Doesn't Need Every Task
Efficient agent benchmarking points to a cheaper way to compare agents: run the tasks that still separate systems, not every task in the suite.
Tool Agents Need State Diffs, Not API Call Scores
Tool-use agents are moving from choosing the right API to changing live product state. That makes a clean function-call score too small a test because...
Healthcare AI Agents Move Beyond Drug Discovery
Healthcare AI agents are moving into admin, triage and prior-authorisation workflows. The real gate is safety, evidence and accountable handoff.
Multilingual Agents Need Workflow Tests, Not Translation Scores
PolyWorkBench landed on arXiv on 7 July 2026 with a useful correction to enterprise-agent hype: a global workflow is not a translated English task...