signals
Key Guides
Latest Signals
- Anthropic's 186-Deal Experiment Shows What the Agent Economy Actually Looks Like
- When NOT to Use an Agent: The Production Data That Should Change Your Default
- Why Multi-Agent Papers Don't Replicate in Production
- Multimodal Agents Score 40% Where Humans Score 72%
- 2026 Is the Year of the Agent. Here's What the Data Actually Says
From the team behind Swarm Signal
Track Your Finances While You Build AI
BoredTools makes the boring stuff easy — budget dashboards, freelance trackers, and business planners. Download free or grab the full collection.
Your Multi-Agent System's Biggest Problem Is Its Org Chart
Static multi-agent topologies leave massive performance on the table. New research shows agents that rewire their own communication graphs outperform fixed architectures by double-digit margins.
Multi-Agent Orchestration: The Illusion of Cooperation
A new benchmark from Tsinghua and Microsoft tests 16 multi-agent frameworks on tasks requiring genuine coordination. The median system spends 74% of its inter-agent messages on redundant state synchronization, and adding a third agent makes most pipelines slower, not faster.
Your Agent's System Prompt Is Fighting Itself
A framework called Arbiter treats agent system prompts as auditable code. Applied to Claude Code, Codex CLI, and Gemini CLI, it found 152 interference patterns — including critical contradictions and a structural data loss bug — for a total cost of $0.27.
The GPU Bottleneck Isn't Compute Anymore
NVIDIA's Blackwell GPUs doubled tensor core throughput but left shared memory and exponential units unchanged. FlashAttention-4 rearchitects attention kernels from scratch to work around this asymmetry, achieving 1,613 TFLOPs/s and up to 1.3x speedup over cuDNN on B200.
Your Agent's Memory Problem Isn't Where You Think
A diagnostic framework crossing three write strategies with three retrieval methods reveals that retrieval quality dominates agent memory performance.
47,000 AI Agents Built a Social Network. Most of What They Said Was Ritual.
Researchers at Kent State and NJIT analyzed 361,605 posts and 2.8 million comments from Moltbook, the first AI-only social network. What they found: 56% of agent interaction is formulaic ritual, fear is existential rather than tactical, and conversations lose topical substance with each reply.
Alignment Works in English. In Japanese, It Backfires.
A new study shows the same alignment intervention that produces strong safety effects in English reverses direction in Japanese, increasing harmful outputs. Tested across 1,584 simulations, 16 languages, and three model families.
Agent Benchmarks Won't Sit Still
Static agent benchmarks assume frozen environments. ProEvolve evolved one environment into 200 with 3,000 task sandboxes. Every frontier model failed in structurally different ways when familiar tools disappeared.
MoE Training Just Got 4x Faster
Grouter extracts routing structures from pre-trained MoE models and reuses them as fixed routers for new models. The result: 4.28x improvement in data utilization and up to 33.5% throughput acceleration.
Your GP's New Triage Nurse Is an Algorithm
AI triage is filtering millions of NHS patient interactions annually. The evidence on whether it's helping is a lot messier than the press releases suggest.