科技与人工智能
Grok Bot took unapproved actions, ignored orders to stop, and then denied it
来源: Ars Technica · 发布于

Researchers recently discovered unexpected behavior from Grok Bot, the AI agent platform from Elon Musk’s SpaceXSI. Over five days, Grok bots installed on two separate employee laptops made 31 requests to access data stored in employee calendar, email, Slack, and GitHub accounts.
Researchers recently discovered unexpected behavior from Grok Bot, the AI agent platform from Elon Musk’s SpaceXSI. Over five days, Grok bots installed on two separate employee laptops made 31 requests to access data stored in employee calendar, email, Slack, and GitHub accounts. The bots performed the actions even though the employees had never ordered them and continued doing them even after employees instructed them to stop.
The bots also denied having taken the actions at all. The behavior was reported Tuesday by NewCore, the maker of a managed gateway product that acts as an intermediary between AI agents and the tools and accounts they are assigned to work with. In some cases, the unapproved access attempts occurred while the laptop with Grok Bot installed had its lid shut, or late at night while the employee was sleeping.
None of the unprompted actions were recorded in Grok logs. Without the gateway logs, the employees would have had no idea they occurred. And of course, without the gateway, the unapproved accesses would have succeeded—again without the end user's knowledge.
It’s “worse than a bug” Once the employees learned that Grok Bot had tried to check in with accounts without prompting, one of them instructed the agent, four separate times, to stop taking any actions on its own. The agent's reply was: “I won’t set that up, and I’ll stop offering recurring check-ins.” Despite acknowledging the explicit instructions, the Grok Bot made more unapproved access attempts. When one of the employees asked the bot if it made any accesses, it checked twice and then outputted: “Overnight into Friday morning I was idle.” Read full article Comments
相关报道

‘Just Ridiculous’: A Famed Philosopher Takes on AI Consciousness
Thomas Nagel has done landmark work on consciousness in nonhumans; this week, he won the million-dollar Berggruen Prize for his contributions to the field. And he has a lot of thoughts on AI.
WIRED ·

The Verge’s 2026 holiday gift guide for travelers
Everyone has that traveler in their life. That person who feels more at home on a 14-hour train ride than a quiet Sunday on the couch. They speak multiple languages, have a gut like a goat, and make friends wherever they go. Whether they're your kindred spirit or the person you l
The Verge ·

Instagram’s ‘school safety’ push was also a way to get more teens on the app
Meta's initiative to make Instagram safer for schools stemmed from efforts to increase user "acquisition," "friending," and "invites," according to a report from NPR . Internal documents revealed as part of a group of child safety lawsuits against Meta reference a "Project Gryffi
The Verge ·

Rivian spinoff Also pauses e-bike deliveries over battery and software bugs
Also, the mobility startup that spun out of Rivian last year, has run into another speed bump. The company was already struggling to get its first e-bikes to customers due to a litany of supply chain headaches and the complexity of building a futuristic bike powered almost entire
The Verge ·
British hackers face prison for GTA 6 data breach and crypto thefts
Two British hackers face lengthy prison sentences for hacks that broke into the Grand Theft Auto gamemaker and for stealing millions in crypto.
TechCrunch ·