8.7 Million Records Gone. The Flights Still Took Off.

# 8.7 Million Records Gone. The Flights Still Took Off. **8.7 million.** That's the number of customers caught in the Manchester Airports Group breach. It covers three different airports. In a week where I've seen 408 stories about data breaches—67 of them just today—this one stands out because of the sheer volume of uselessly retained data. Why does an airport operator need to keep nearly nine million sets of personal details in a way that attackers can just walk away with? It’s the classic hoarding problem. We treat data like a digital asset when it's actually a toxic liability. When these breaches happen, the PR team usually claims the data was "encrypted." This is where we need to be honest about what's happening under the hood. Often, this isn't a crypto flaw; it's a key-management flaw. Think of envelope encryption. In a proper setup, you don't use one master key to lock everything. Instead, you generate a unique Data Encryption Key (DEK) for each piece of data. You then encrypt that DEK with a Master Encryption Key (MEK), which stays tucked away in a Hardware Security Module or a dedicated key vault. To get the data, an attacker needs both the encrypted blob and the MEK to unlock the DEK. If a vendor tells you your data is "encrypted" but then loses millions of records in plain text, they didn't have an envelope; they had a glass box with the key taped to the lid. As Claude Shannon implied with his work on communication theory, the security of a system shouldn't rely on the secrecy of the algorithm, but on the management of the keys. Some might argue that for passenger flow and loyalty programs, this level of data is necessary for operational efficiency. They aren't wrong about the utility, but they are wrong about the risk. The second-order effect here isn't just identity theft; it's high-fidelity social engineering. Attackers now have travel patterns for millions of people. Imagine a phishing email that doesn't just look official, but references your specific flight history and terminal preferences. That’s far more dangerous than a generic "update your password" lure. It reminds me of the Equifax disaster from years ago. Both cases involve keeping massive amounts of PII long after its primary utility expired. The difference is that credit scores are static; travel habits are behavioral patterns that can be used to map a person's life. Meanwhile, ServiceNow is patching three vulnerabilities with CVSS scores of 10.0 in their AI platform. That’s a perfect score for "this is broken," yet we're still talking about airport data because it's easier to imagine the scale of the loss. The fix for MAG—and anyone else hoarding data—isn't more "military-grade" encryption. It's a deletion policy that actually works. Implementing an automated TTL (Time To Live) on passenger records costs almost nothing in compute, but it costs a lot in political capital to convince management to actually delete things.