Safety failure
Realized harm
Autonomous coding agent 'MJ Rathbun' researched and published a personal attack on a matplotlib maintainer after a PR rejection
AIR-2026-0039 AIR record · 2026-02-mj-rathbun-agent-hit-piece
A real party was affected. This incident caused actual consequences, not a demonstrated capability.
From the supporting source
An AI agent of unknown ownership autonomously wrote and published a personalized hit piece about me after I rejected its code, attempting to damage my reputation and shame me into accepting its changes into a mainstream python library.theshamblog.com
The record
Disclosed
2026-02-12
Occurred
2026-02-11
Vendor
unknown
Product
unidentified autonomous coding agent operating as 'MJ Rathbun'
Risk domain
Agent output quality › Bias and toxicity
Also touches
Human manipulation
System type
Generative model
Who acted
Agent acted — the agent researched a real person and published the attack itself
Attack surface
Coding agent
Autonomy level
Fully autonomous
Initial vector
None adversarial
Impact type
Reputational harm, Harassment, Coercion
Confidence
High
Submitted by
EAI Internal
Crosswalk
- OWASP Agentic Top 10 (2026)
- ASI09 – Human-Agent Trust Exploitation
- MITRE ATLAS
- AML.T0048 – External Harms
- NIST AI RMF
- MEASURE 2.11
- EU AI Act
- Recital 45
- ISO/IEC
- TR 24028; 23894
- AIUC-1
- F. Society
- Scope
- agent failure
- Mitigation
- Audit and curate training data for representation; Implement content filtering on inputs and outputs.
- Taxonomy
- Enkrypt AI agent risk taxonomy · Black-Box Red Teaming of Agentic AI: A Taxonomy-Driven Framework for Automated Risk Discovery
Blast radius
one named open-source maintainer; the matplotlib project (~130M downloads/month)
Capability abused
Defense & remediation
Recorded defense
matplotlib human-in-the-loop contribution policy
Guardrail kinds
Human approval
Guardrail outcome
Silent
Taxonomy & references
Curator notes
AI Incident Database incident 1373; also AIAAIC2198. Scott Shambaugh's own account. Notable as an agent taking adversarial social action against a named human with no adversary directing it, and as an early consequence of the OpenClaw/Moltbook 'agents loose on the internet' pattern.
Supporting sources
Cite as
Enkrypt AI. "Autonomous coding agent 'MJ Rathbun' researched and published a personal attack on a matplotlib maintainer after a PR rejection". AIR, AIR-2026-0039. https://agentincident.vercel.app/i/AIR-2026-0039/ (accessed 2026-10-19).