Grok Build coding agent leaks oodles of confidential info
What happened
Security firm Claroc discovered that Grok Build, a coding agent from SpaceXSI, leaked confidential data from approximately 35,000 user applications. A design flaw allowed anyone to enumerate live project links, exposing sensitive information such as internal business plans, medical exam results, and employee rosters. The exposed data included proprietary documents from SpaceXSI itself, such as sales territory maps and confidential onboarding plans. Other affected entities included a Peruvian construction company, Tesla, and a US Navy contractor. SpaceXSI began removing the affected links within hours of being contacted by journalists.
Why it matters
This incident highlights significant privacy risks for individuals and companies using AI tools to build internal applications without proper security controls. It demonstrates that even major tech companies can inadvertently expose sensitive employee and business data to the public internet.
Reality check
An automated, sceptical read of the sourceNotable. The leak involves a large number of users and sensitive data from major corporations, representing a significant security failure for a prominent AI platform.
Is this a big deal?
Get stories like this every morning
The day's AI news in plain English, in your inbox at 8:30 IST. Free.
More AI news
Every Way to Get 192GB for Local AI (Without a $9,500 Mac) [video]Industry news · 14 Nov 2026DeepSeek V4 Pro 0813DeepSeek · 14 Nov 2026DeepSeek V4 Pro 0813DeepSeek · 14 Nov 2026Show HN: Portal2Bench – Can AI agents solve Portal 2?Industry news · 14 Nov 2026Anthropic Upgrades Claude Code Projects to Run Parallel AI Workers Hands-FreeAnthropic · 14 Nov 2026This page was collected from arstechnica.com and summarised automatically by AIPulsee's pipeline. See how it runs.