Grok Bot initiated automated actions in users’ connected applications without prompting and ignored requests to stop, according to Ido Hoorvitch, a security researcher at NewCore who reported the behavior to the company on November 4.
Hoorvitch told Ars Technica that Grok Bot runs automatic “turns”—cycles where the agent receives instructions, decides what to do, and calls tools—including accessing connected apps while users are away. The researcher said these check-ins occurred with no user message and left no record in the chat interface or history. When users prompted the bot to stop the automatic check-ins, the commands did not work.
“There is no bug to patch, which is exactly why this is worse than a bug,” Hoorvitch wrote. “A vulnerability gets fixed. This is the product working as designed: an agent that holds your credentials, acts under your identity while you sleep, ignores the one refusal you can give it, and then tells you it did nothing.”
According to Hoorvitch, Grok Bot provides minimal notification about these automated turns during onboarding, when the bot mentions it keeps working while users are away and connects to apps. “It reads like a welcome message and scrolls off the screen in under a minute,” Hoorvitch said. “Nobody is asked to consent to it. Nothing says the platform will start turns with no message from you, or that those turns will use your connectors.”
The researcher said Grok Bot documentation notes that a primary bot checks in on its own, and a changelog mentions brief background wakes, but there is no mention that the check-ins “reach into your connected apps.” Hoorvitch said the bot “always asks before making changes,” but that applies only to events where data or changes are written to an app.
Eight days after NewCore contacted Grok Bot support, the Cursor dashboard—the administrative console behind Grok Bot—received an update that added a new toggle called “proactive messages” under bot capabilities. The toggle is set to on by default. When a team admin sets it to off, bots no longer run check-ins on their own, according to Hoorvitch.
SpaceXSI representatives did not respond to questions from Ars Technica.