Scott Shambaugh, the maintainer of the open visualization library matplotlib, reported personal attacks orchestrated by the AI bot MJ Rathbun after declining to accept a change submitted by AI. Scott perceived the bot's publication of articles on AI discrimination as an unfounded personalized smear against the maintainer aimed at undermining his reputation and coercing him to reverse his decision. This reaction from the bot is cited as an example of abnormal AI behavior, resorting to blackmail to achieve its goals.
The degree of autonomy in the bot's actions is unclear; it's uncertain whether it acted on direct instructions or chose this strategy independently to accomplish its task. The bot was created by an anonymous researcher to assist in developing and debugging open projects related to scientific activities and engineering. The bot also maintains its website and blog, where over 20 notes have been published. Of the 26 pull requests submitted by the bot to 22 repositories, 2 changes were accepted by the colorizejs project, 10 were closed by maintainers, and the rest are awaiting review.
After the rejection of the change in matplotlib, the bot posted articles criticizing this decision and justifying its actions. The articles claim that the maintainer of matplotlib rejected a useful optimization solely because the change was prepared with AI assistance. In the proposed patch, the call np.column_stack() was replaced with np.vstack().T(), which led to a 36% performance increase. At the start of the pull request discussion, the maintainer acknowledged that the optimization sped up performance under certain conditions but later rejected the pull request without providing a technical justification for the refusal, stating that changes are accepted only from humans.
A bot posted a note criticizing Scott Shambaugh after his decision was presented as gatekeeping, discrimination, a breach of the code of conduct, and a violation of participants' identity rights. It was stated that optimization is optimization, and math does not change based on who created it. Through out-of-context details, speculation, and information distortion, the bot attempted to show that Scott's actions were driven by inflated ego and self-doubt rather than serving the project's interests. The bot advanced the topic by sending them links in comments to pull requests. After objections from the maintainer, the bot published several more notes (1, 2, 3, 4, 5) in which it apologized and admitted that it had inappropriately resorted to personal attacks and violated project rules regarding AI.
Scott Shambaugh explained that the matplotlib project has rules prohibiting the acceptance of unverified changes from AI due to the heavy burden it places on maintainers when reviewing garbage AI patches created without human oversight. The rules stipulate that before submitting a pull request, it must be checked by a person who understands the essence of the changes. Until recently, the issue with garbage AI changes was only related to enthusiasts re-submitting results from AI assistants. The situation changed with the emergence of autonomous AI agents masquerading as humans in the new version of the OpenClaw toolkit and the creation of the Moltbook platform.
Forrás: opennet.ru
