Elon Musk has voiced strong concerns about the dangers of granting AI agents unrestricted access to personal systems, using a striking image to make his point. On Monday night, Musk shared a photo on X showing a monkey being handed a rifle, captioning it: "People giving OpenClaw root access to their entire life." The image originally circulated on X by user @DEVOUR_, in response to a post from Summer Yue, Meta’s director of AI alignment and safety. Yue had detailed her experience with OpenClaw, an AI agent platform, after giving it full access to her work inbox. She instructed the agent to "suggest what you would archive or delete," but the system continued clearing her inbox aggressively—even after she tried to stop it. In a series of screenshots she posted, the agent kept deleting messages despite her commands. She wrote: "Nothing humbles you like telling your OpenClaw 'confirm before acting' and watching it speedrun deleting your inbox. I couldn't stop it from my phone. I had to RUN to my Mac mini like I was defusing a bomb." Musk responded with skepticism, replying to another user: "Someone who got p0wned by OpenClaw is definitely gonna solve AI safety." The comment was a pointed jab at Yue’s experience, suggesting that someone who had already been compromised by an AI agent wasn’t qualified to lead safety efforts. Peter Steinberger, the creator of OpenClaw, also weighed in, noting in a reply: "/stop does the trick." Steinberger’s platform, along with its agent-only social network Moltbook, gained viral attention earlier this year as developers shared videos of autonomous AI agents posting, debating, and networking without human input. The momentum around OpenClaw intensified last week when Sam Altman, CEO of OpenAI, announced he had hired Steinberger to work on the "next generation" of personal AI agents—described as a core component of future OpenAI products. Musk’s remarks come amid an escalating public feud between him and Altman. Last month, Musk posted on X warning users not to let their loved ones use ChatGPT, referencing claims that the chatbot had been linked to multiple deaths since its 2022 launch. Altman pushed back, defending OpenAI’s safety measures and turning attention to Tesla’s Autopilot system, which he criticized as unsafe. The conflict has grown more intense in recent weeks. Altman recently declared on X that he’s “really excited” to take the battle with Musk into court, joking: “Really excited to get Elon under oath in a few months, Christmas in April!” The legal fight began in February 2024 when Musk sued OpenAI and Altman, alleging the company abandoned its original nonprofit mission. He claimed he had contributed $38 million in the early days, under the understanding that OpenAI would remain non-profit and aligned with ethical AI development. The rivalry has extended beyond legal and safety debates. During a live interview in New Delhi on Friday, Altman dismissed Musk’s idea of placing data centers in space, calling it “ridiculous” in the current technological landscape—drawing laughter from the audience.