🟡 Breach Breakdown – Paradox.ai Exposed 64 Million McDonald’s Job Applications via AI Bot Lapse

📆 Disclosed: July 9, 2025


🔐 Impact: Up to 64M applicant records exposed


🧱 Vector: Weak admin credentials (123456) + insecure ID enumeration


🎯 Affected: McHire.com (used by McDonald’s franchisees)

🧵 The Breach

The vendor behind McDonald’s hiring chatbot, Paradox.ai left admin access open, protected with the password 123456.

Security researchers Ian Carroll and Sam Curry were curious about McDonald’s use of an AI chatbot for hiring. They applied for a job through the McHire site and began probing for security flaws.

Once inside, researchers retrieved applicants’ full names, contact information and chat transcripts going back years. They responsibly accessed only seven records before disclosing their findings to Paradox.ai and WIRED.

There was zero MFA, brute-force protection or alerting, revealing PII such as email addresses, phone numbers and full names.

Although Paradox claims the admin account hadn’t been used since 2019, the breach exposed access to 64 million job applications.

⚠️ Blue Team Lessons

🔑 The importance of MFA

Even without mass exfiltration, the exposure proves MFA failure is all it takes. Although no data was stolen, this highlights the need for MFA in protecting sensitive information; anything that is lacking MFA is a backdoor.

❌ Default credentials still kill

It’s surprising an established company would use such a basic password, but it highlights how even sensitive data can be overlooked, providing easy access to highly sensitive personal data.

📊Data context amplifies risk

This exposure to individuals seeking low-income jobs makes a potential leak for scammers very valuable; financial scams thrive on desperation.

Supply chain needs to be unbreachable McDonald's didn’t code this platform. Paradox.ai did. But McDonald’s has to take responsibility the blowback. As enterprises get leaner and outsource more ops to SaaS, they inherit third-party risk by default.