In a recent development, security researchers found that McDonald’s AI hiring chatbot was left wide open, exposing information from more than 64 million applicants.
This isn’t the first time McDonald’s technology has faced criticism for bugs, vulnerabilities, and breaches. However, this incident should teach us a valuable lesson on the real-world risks that exist when AI chatbots are used to hire new employees.
Let’s explore what happened and why even resourceful and mainstream companies like McDonald’s fail to get AI security right.
Bug bounty researcher pentests the McHire AI bot system and gets instant access
During a penetration test, Ian Caroll, a side-hustling cybersecurity researcher and bug bounty hunter, found that the McHire — the AI chatbot used by 90% of McDonald’s franchisees to hire new employees — exposed the data of more than 64 million applicants.
The researcher started looking into McHire after reading posts of Reddit users who complained about the app and its erratic behavior.
Caroll started looking into McHire and, shockingly, found that the entire database of McDonald’s new hires, along with data from franchise companies, could be accessed using default credentials and passwords, specifically 123456:123456.
Yes, read that correctly. The password for the entire system was 123456.
This critical vulnerability would be hilarious if it didn’t put the over 64 million people who used the chatbot to apply for McDonald’s employment at risk of phishing and other cyberattacks.
What information was exposed by McDonald’s AI hiring bot?
According to the research, the information exposed by the McDonald’s AI hiring bot includes applicants’ personal data, including their personality test results, full addresses, phone numbers, CVs, and more.
While we cannot confirm it, it is possible that this data had already been leaked on the dark web and spread through messaging channels like Telegram. This type of information is used in bulk by cybercriminals when creating more convincing automated phishing and spear phishing campaigns.
These campaigns usually target victims for identity theft, account takeover, and the spreading of malware that steals financial and personal information.
The McHire bot collects applicants’ personal information, shifts preferences, and administers personality tests.
“We could see all of the employees of the restaurant were simply employees of Paradox.ai, the company behind McHire,” Caroll said. “Entire chat history with the bots, phone number, email, shifts they want to work in, full address, resume or CVs, and other notes.”
Additionally, the investigation shows that the Auth token to log into the consumer user interface leaked raw chat messages and, presumably, other information.
Screenshots of the penetration test investigation shared by Caroll show that access to the system using the default and insecure 123456 password also seems to expose the following information fields:
- Employee role and job title
- Franchise staff
- Franchise owners
- Company admin
- Supervisors
- Recruiter
- Hiring managed
- Full User – Edit Everything
- And more
I applied to McDonald’s via McHire. What should I do?
We advise that all users who used McHire run free dark web monitoring scans on sites like Have I Been Pwned and remain vigilant for any security warning messages coming from their accounts. If you do get any warnings from your accounts, change your password immediately and enable MFA (if possible, biometrics).
Get used to AI interviews — and being hired and fired by AI bots
McDonald’s is far from the only company using AI in their recruitment, hiring, and onboarding processes. AI in HR has skyrocketed to new heights, with workers often finding that their applications have been screened by AI. It has also become common for workers to have to do video interviews with AI bots before they can even speak to a human.
A recent report found that most HR managers are using AI for various tasks within the department, with 65% using AI to hire and fire workers. The report also found that two-thirds of those using AI in HR admit to having never received any type of training on how to manage people with it.
The use of AI in HR brings numerous benefits for companies, such as the reduction of costs and the scaling, automating, and accelerating of hiring processes. For those looking for work, however, it’s a different story.
Given the speed at which industries are embracing AI hiring and interviewing bots, we expect that the number of companies using this type of technology will be much higher than 65% and growing by the day.
Tips for job applicants who encounter AI on their HR journey
If you’re applying for jobs through AI-powered systems, it’s crucial to stay alert, protect your data, and understand your rights as an applicant. Here are 5 simple tips to keep in mind:
- Use strong, unique passwords when creating accounts on hiring platforms.
- Avoid oversharing. Only provide information that is strictly necessary for your application.
- Check to see how long companies store your data and whether they share it with third parties.
- Monitor your email and other accounts for unusual activity after submitting job applications.
- Whenever possible, ask for a human contact or an alternative way to apply.
Here are 5 additional outside-the-box tips for when you encounter AI bots:
- Use a secondary email address just for job applications. That way, if your data is leaked, your primary inbox remains safe.
- Obfuscate sensitive data slightly on your CV (e.g., write your birthdate as “MM/YYYY”) and only provide full details later in the process.
- Search the platform or chatbot’s name + “breach” or “vulnerability” to see if it has a bad track record before applying.
- Screenshot your chatbot conversations. They might be helpful if you need to report a problem or follow up on your application.
- Ask in online forums (e.g., Reddit, Blind) if others have had strange experiences with the company’s hiring system. Often, red flags appear before they hit the news.
Final thoughts
Every day, more companies turn to AI bots to hire workers, making it likely that in the near future, you will encounter an AI bot in some type of HR setting.
Ultimately, it is up to each individual to take a responsible approach to AI when applying for jobs. AI bot interviews and AI selection processes should not be mandatory, and employees and new candidates should have the right to speak with a human or decline being interviewed by AI or giving their personal data away.
While declining to use an AI in the hiring process may well translate to not being able to apply to a particular job, sometimes — like in the 123456 McDonald case — the risks can outweigh the rewards.
Bottom line: AI in HR is already here. Don’t expect companies to safeguard your data for you. As we have seen, even the biggest and most resourceful enterprises, like McDonald’s, struggle to get AI security right.
This is an independent publication, and it has not been authorized, sponsored, or otherwise approved by McDonald’s Corporation. McDonald’s and McHire are trademarks of McDonald’s Corporation.
