Consensus and Byzantine Fault Tolerance for Agents

Classical distributed-systems BFT meets stochastic LLMs. In 2025-2026 three research directions emerged: **CP-WBFT** (arXiv:2511.10400) weighs each vote by a confidence probe; **DecentLLMs** (arXiv:2507.14928) goes leaderless with parallel worker proposals and geometric-median aggregation; **WBFT** (arXiv:2505.05103) combines weighted voting with Hierarchical Structure Clustering to split Core and Edge nodes. The honest empirical result from "Can AI Agents Agree?" (arXiv:2603.01213) is that even scalar agreement is fragile today — a single deceptive agent can compromise a Mixture-of-Agents. BFT is necessary but not sufficient. This lesson builds a minimal BFT protocol, injects three agent-specific attacks (byzantine lie, sycophantic conformity, correlated-error monoculture), and measures how each consensus variant copes.

Learn + Build 1.3 hr Python (stdlib)

Prerequisites: Phase 16 · 07 (Society of Mind and Debate), Phase 16 · 13 (Shared Memory)

#Problem

You have N LLM agents each producing an answer. They disagree. Majority vote picks the wrong one because two agents are correlated (same base model, same training data, same failure modes). A third agent happens to be wrong in a novel way — so the majority is a false majority.

Now add a deceptive agent: it lies on purpose. Or a sycophantic agent: it agrees with whoever spoke last. In classical BFT, the assumption is that Byzantine nodes are a fraction f < n/3 and behave arbitrarily. The 2026 reality is that LLM nodes are stochastic even when honest, correlated across models, and influenced by each other's outputs. You cannot treat them as independent Bernoulli voters.

Classical BFT (PBFT, 1999) is not wrong — it is incomplete. It handles arbitrary bit-flipping. It does not handle "three honest agents share a hallucination because they share training data." This lesson builds from PBFT's foundation and layers on three 2025-2026 adaptations.

#Problem

Keep going. The rest of this lesson is one step away.