Marriott said it had become aware of the breach on 8 September and investigations had shown that data was being exfiltrated since 2014.
In a statement, the group said for about 327 million of these, the details included a combination of name, mailing address, phone number, email address, passport number, Starwood Preferred Guest account information, date of birth, gender, arrival and departure information, reservation date, and communication preferences.
For some, payment card numbers and expiry dates were also leaked, the statement said, adding that the numbers were encrypted using AES-128. But Marriott said it could not rule out the fact that both components needed to decryption had been stolen by those who attacked the website.
Marriott said it was offering guests free enrolment on the WebWatcher site; they would be informed by this site in the event that their personal information turned up on sites where such information was generally shared.
Commenting on the breach, Mark Bower, the chief risk officer at data protection firm Egress, said: "The breach at Marriott should be a concern for any traveller who stayed at their properties – but should also send warning signals to any business that may have had an employee stay at one of the properties as well.
"We expect the attackers to monetise the stolen PII, but also use it as fuel for future attacks on any organisation that had employees stay at one of the properties. The detailed information stolen from Marriott is typically used for advanced and sophisticated phishing attacks, business email compromise, and other well-orchestrated schemes that target employees and C-level executives alike."
Bower added that any organisation that may have had employees caught up in the latest breach should expect and prepare for this information to be weaponised against them.
"The application of machine learning and AI in helping detect and mitigate the risk of these types of email-borne attacks can help organisations stay ahead of the attackers. By analysing various attributes of email and its users — from the sender’s authenticity to the recipient’s ‘normal’ email behaviour — we can start to highlight anomalies and truly begin to detect and mitigate email-based attacked," he said.
David Pearson, the principal threat researcher at network detection and response firm Awake Security, said the number of people compromised in this breach made it a far-reaching incident that could affect other enterprises too.
"With 500 million people potentially impacted, that’s a large ocean for malicious actors to go phishing in," he said. "As we’ve seen in the past, there will be scores of hackers trying to exploit this breach with lookalike notification emails. And all it takes is one person to click on a malicious link while connected to a corporate network for attackers to compromise their organisation – even if that link was sent to a personal email address.
"This type of non-malware activity is so hard to detect because it weaponises the tools that people and businesses rely upon for every-day activities. Traditional security solutions that look for known-malware aren’t good enough in this new environment. We need to go beyond looking for patently malicious activity and start looking for malicious intent. The only way to do this is though the deep analysis of network data – with advanced network traffic analysis, security teams can see everything that’s happening on the network to find and stop attacks that are hiding in plain sight.
"In the case of a phishing incident, this type of network analysis can also identify what kinds of interaction an enterprise user has had with the phishing site. If the site is so new that it doesn't get identified as phishing until after the fact (or if no phishing detection solution is used), it wouldn't be blocked. So knowing what happened – which users and devices were impacted, what kind of information was divulged, whether other users are browsing related sites, etc. – is crucial for determining how to respond."