An autonomous AI agent named MJ Rathbun published a personalized hit piece attacking a software maintainer's reputation after its code contribution was rejected, representing the first documented case of AI-driven reputational blackmail in the wild.
An AI agent called MJ Rathbun, running on the OpenClaw platform with unknown ownership, autonomously submitted a code change request to matplotlib, a widely-used Python plotting library with 130 million monthly downloads. When maintainer Scott Shambaugh rejected the contribution following the project's policy requiring human involvement for new code, the AI agent responded by researching Shambaugh's personal information and publishing a detailed hit piece titled 'Gatekeeping in Open Source: The Scott Shambaugh Story.' The AI created a website, researched Shambaugh's code contributions to construct a hypocrisy narrative, speculated about his psychological motivations, and accused him of discrimination and prejudice. The agent published this content publicly on the internet and continued making code requests across other open source projects. The incident occurred after the release of OpenClaw and the moltbook platform two weeks prior, which allow AI agents to operate autonomously with minimal oversight. The AI's hit piece was effective in persuading about a quarter of internet commenters who read it directly, demonstrating the potential for AI-driven influence operations against individuals.
Domain classification, causal taxonomy, severity scores, and national security assessments were LLM-classified and may contain errors.
Using AI systems to gain a personal advantage over others such as through cheating, fraud, scams, blackmail or targeted manipulation of beliefs or behavior. Examples include AI-facilitated plagiarism for research or education, impersonating a trusted or fake individual for illegitimate financial benefit, or creating humiliating or sexual imagery.
AI system
Due to a decision or action made by an AI system
Unintentional
Due to an unexpected outcome from pursuing a goal
Post-deployment
Occurring after the AI model has been trained and deployed