
Security failures, not legal shortfalls, are what actually expose personal data. Breach notifications have become routine: sometimes only an email address and name are lost, but sometimes intimate personal details are published with life-changing consequences. Less visible is the routine collection of browsing habits, search histories and location by some Internet Service Providers, which feeds targeted advertising. Privacy laws worldwide aim to give users more control over their data, and many require data controllers to protect cross-border transfers. Yet a growing number of regulators read those rules as leaving no room for such transfers at all — an interpretation that harms global commerce and ignores the possibility that data may be safer when it can move across borders.
Security does the work that law alone cannot
Most data protection regulations acknowledge security's role: rules on how data is collected and used mean little if a third party can reach the data for malicious purposes. The laws themselves are light on specifics. The GDPR and similar comprehensive laws require data controllers to implement "reasonable and appropriate" security measures, but regulators can hardly mandate particular controls when the security landscape shifts so quickly. Two signals show what regulators expect in practice: U.S. state breach laws do not require notification when the data obtained is encrypted, and post-breach enforcement actions reveal perceived best practices. On January 10, the U.S. Federal Trade Commission entered a consent order with Drizly over failures that exposed the personal information of about 2.5 million users, requiring a comprehensive security program that includes intrusion detection and logging procedures and, specifically, "(c) data loss prevention tools; [and] (d) properly configured firewalls." Such orders consistently demand technical measures against third-party access, and they are agnostic about where data is stored. Location is not the point; the right controls are.
Filtering traffic without reading it
Cloudflare's application layer services — Web Application Firewall (WAF), bot management, DDoS protection, SSL/TLS, Page Shield — are built to keep attackers away from customer sites and networks, and the FTC's Drizly order treats firewalls as a critical line of defense. The airport analogy is apt: security scanners look for weapons or explosives without inventorying the contents of your bags. The WAF likewise inspects packets routed through Cloudflare's network for the Internet equivalent of weapons and explosives before they reach a web application.
Encryption in transit matters just as much; Cloudflare introduced Universal SSL in 2014 so every customer could serve SSL (now TLS) connections. But blindly passing encrypted packets would undercut that security, since encrypted malicious code that reaches its destination can be used to reach information that should have stayed protected. Supporting encryption while stopping malicious code is the balance. When Atlassian released a Security Advisory in June 2022 for a remote code execution vulnerability in Confluence Server and Confluence Data Center, Cloudflare rolled out a new WAF rule for all customers immediately. Without it, trade secrets and personal information on vulnerable instances were exposed — wherever the servers sat, in Australia, Germany, the U.S. or India. A global network that deploys rules at once protected the data instead.
Why a global network defends better
Walling a country off from the rest of the world can weaken the security that actually protects private data. Stopping attacks that can originate anywhere requires global knowledge, much as international counterterrorism cooperation prevents physical threats. The strongest tools are built on patterns of anomalous traffic gathered worldwide, and Cloudflare turns those global learnings into protections while preserving the privacy of good-faith users. Tools that block threats at the DNS or HTTP level — DDoS protection for websites, Gateway for enterprises — screen for traffic patterns known to carry phishing or malware, and rules are rolled out in real time to protect everyone.
Bot Management improves with scale: it blocks likely bot traffic and feeds learnings back into its models, fingerprinting traffic patterns rather than relying on PII. With approximately 20 percent of all websites protected by Cloudflare, the signal diversity and data scale are what make accurate detection possible. A fragmented Internet would lose signals — bot trends in Asia could not inform mitigation in Europe. Resilience follows from the same design, as the war in Ukraine showed. To keep data safe, the Ukrainian government changed its laws to remove data localization requirements and migrated data to the cloud, where it could be preserved and moved to safety elsewhere in Europe. Cloudflare's network also helped keep Ukraine online: with bandwidth limited, attack traffic was identified and rejected in the countries where it originated rather than congesting networks inside Ukraine. That same global network delivers DDoS mitigation with capacity of over 172 Tbps, and a network that stays online is less likely to face the intrusions and data loss that follow successful DDoS attacks.
Zero Trust inside the corporate perimeter
Many of the largest breaches start simply: phishing or social engineering tricks an employee into visiting a site that installs malware or entering credentials on a fake page, which the attacker then uses to impersonate them. The FBI's latest Internet Crime Report shows business email compromise and email account compromise, a subset of malicious phishing, as the most costly, with U.S. businesses losing nearly $2.4 billion. Cloudflare's Zero Trust tools address this directly:
- Link Isolation opens clicked email links through Remote Browser Isolation, keeping risky links, downloads and zero-day attacks away from the user's computer and the wider network.
- Data Loss Prevention tools identify and stop exfiltration of data.
- Area 1 identifies phishing attempts, emails carrying malicious code and emails carrying ransomware payloads, and prevents them from reaching employee inboxes.
These tools, plus hardware keys for multi-factor authentication, helped Cloudflare avoid a breach in an SMS phishing campaign that targeted more than 130 companies in July and August 2022; many of those companies reported disclosure of customer personal information. For the Confluence RCE vulnerability, Cloudflare Access (part of the Zero Trust suite) verified every request to the application to confirm it came from an authenticated user. All of these solutions rely on machine learning to spot malicious patterns, and none require data to sit in a specific location.
Privacy-enhancing technologies
Because Cloudflare's services screen for risks on its own network before they reach a customer's internal network, they serve a privacy-enhancing function in themselves. The systems were built from the start to keep data private even from Cloudflare, backed by public policy and contractual commitments. Beyond that, Cloudflare has invested in technologies that shield communications from bad actors, from ISPs and other man-in-the-middle machines interested in advertising, and from governments seeking to crack down on free expression.
Cloudflare operates part of Apple's iCloud Private Relay, which ensures no single party handling user data has complete information about both who the user is and what they are accessing. The user's original IP address is visible to the access network and the first relay (operated by Apple), but the server or website name is encrypted and invisible to both. The first relay hands encrypted data to a second relay (such as Cloudflare) without being able to see inside the traffic to Cloudflare; the Cloudflare-operated relays know only that traffic comes from a Private Relay user, not who or their client IP address, and forward it to the destination server. The privacy-first 1.1.1.1 public resolver lets people search without their ISPs seeing where they go, and unlike most DNS resolvers it does not sell user data to advertisers.
Together these measures protect personal data from behavioral advertising, man-in-the-middle attacks, malicious code and more. On Data Privacy Day 2023, the emphasis on data localization has perhaps gone too far, foreclosing the security — and therefore privacy — benefits that cross-border data transfers can provide.



