Security researcher Ian Carroll has disclosed a significant vulnerability in Front Gate Tickets. This Live Nation subsidiary handles ticketing for most major US music festivals, including Lollapalooza, South by Southwest, Austin City Limits and Bonnaroo. The disclosure, first reported by WIRED, is notable because Carroll found and exploited the vulnerability with substantial help from Claude Opus 4.7, Anthropic’s AI model which raises broader questions about how quickly AI can now dig up exploitable bugs across the web.
Carroll first spotted a SQL injection vulnerability on Front Gate’s site which is a common flaw that lets an attacker run commands on a website’s backend. A web application firewall appeared to be blocking him. He asked Claude to find a way past it. The AI on its own came up with a nested SQL query technique that bypassed the firewall. “It was the first time, really, that I had a vulnerability that I didn’t fully understand,” Carroll told WIRED. “I had to go back and read what Claude had written to understand the bypass, because I didn’t write it.”
From there, Carroll was able to access hundreds of databases containing customer and staff data, including names, emails and mailing addresses (though not credit card details) and ultimately take over a super-administrator account by exploiting how the site handled password resets. With that access, he could issue tickets of any value to anyone for any event. He found a Bonnaroo Platinum ticket priced at $4,000 he could add to a cart and duplicate freely. He didn’t complete any orders, flagged the issue to Front Gate instead and the vulnerability was patched within 24 hours.
Carroll is part of Anthropic’s Cyber Verification Program which gives approved security researchers access to Claude for legitimate security research. Anthropic said in a statement that if Carroll had not been part of the program, his use of Claude for this purpose “would have been detected and blocked.”
The broader implication Carroll flags is less about this specific vulnerability and more about what AI makes possible at scale. “I think there’s a very good chance it could have found this exploit end-to-end without me doing anything at all,” he said.