Responsabilité liée au WiFi public : pourquoi le filtrage de contenu est obligatoire

Executive Summary

For IT managers, network architects, and CTOs overseeing public venues, deploying Guest WiFi is a baseline operational requirement. However, providing an open pipe to the internet without robust content filtering exposes the venue to severe legal, financial, and reputational risks. When you provide public internet access, your organisation assumes the role of an Internet Service Provider (ISP). If malicious or illegal traffic — such as copyright infringement, peer-to-peer (P2P) piracy, or Child Sexual Abuse Material (CSAM) — originates from your public IP addresses, the liability often falls on the venue operator.

This guide provides a definitive technical framework for implementing mandatory content filtering. We explore the architecture required to maintain safe harbour protections, ensure regulatory compliance (including GDPR and PCI DSS), and maintain network performance. By integrating robust filtering with WiFi Analytics , venues in Retail , Hospitality , Healthcare , and Transport sectors can mitigate risk while maintaining a seamless guest experience.

Technical Deep-Dive

The Legal Landscape and Safe Harbour

The primary driver for content filtering is public WiFi legal liability. In most jurisdictions, ISPs and public WiFi providers are protected by "safe harbour" provisions — for example, the Digital Millennium Copyright Act (DMCA) in the US, or the E-Commerce Directive and its successor frameworks in the EU. However, these protections are explicitly conditional. To qualify, providers must demonstrate they have taken reasonable technical steps to prevent illegal activity and can assist law enforcement when required.

Without an audit trail and active filtering, a venue cannot prove it took reasonable steps, which nullifies safe harbour protections entirely. This is particularly critical for public sector deployments, where accountability requirements are even more stringent. For context on how public sector digital infrastructure is evolving, see Purple Appoints Iain Fox as VP Growth – Public Sector to Drive Digital Inclusion and Smart City Innovation .

The three primary legal risk vectors for unfiltered networks are:

Architecture of a Filtered Network

Effective content filtering requires a multi-layered architecture. No single control is sufficient. The following layers must work in concert:

Layer 1 — Authentication (Captive Portal): Before network access is granted, users must authenticate. This ties a device (MAC address) and an IP lease to a verified identity via SMS, email, or social login. This is the foundation of your audit trail. For more on why this record-keeping is critical, see Explain what is audit trail for IT Security in 2026 .

Layer 2 — DNS Filtering Engine: The most scalable approach for high-throughput environments is cloud-based DNS filtering. When a user requests a domain, the DNS resolver checks the request against a real-time threat intelligence database. If the domain is categorized as malicious or illegal — malware, adult content, piracy trackers — the resolution is blocked and the user is redirected to a policy-compliant block page.

Layer 3 — Application Layer Gateway (Firewall): DNS filtering alone is insufficient. Users can bypass DNS filters using direct IP connections or encrypted DNS (DNS over HTTPS — DoH). The network gateway must block known DoH resolvers and restrict specific protocols, particularly P2P protocols like BitTorrent, which are the primary vector for copyright infringement on public networks.

Layer 4 — Logging and Audit Trail: All session data — authenticated identity, MAC address, assigned IP, timestamps, and session duration — must be logged securely and retained for the legally mandated period. This data must be accessible to law enforcement on request without compromising other users' data under GDPR principles.

Addressing the DoH Problem

DNS over HTTPS (DoH) is the single biggest technical challenge for content filtering in 2025 and beyond. Modern browsers — including Chrome, Firefox, and Edge — can be configured to use DoH by default, routing DNS queries over HTTPS to resolvers like Cloudflare (1.1.1.1) or Google (8.8.8.8). This completely bypasses your managed DNS filtering layer.

The mitigation strategy has two components:

1. Blocklist known DoH resolver IPs at the firewall level. Maintain an updated list of known DoH endpoints and block outbound HTTPS traffic to those specific IPs.
2. Intercept and redirect all port 53 traffic to your managed DNS resolver using firewall NAT rules, preventing manual DNS override by guests.

Implementation Guide

Deploying a robust filtering solution requires careful planning to balance security with user experience. The following steps apply to venues of all scales, from a single-site hotel to a multi-location Retail chain.

Step 1: Define the Acceptable Use Policy

Establish a clear Acceptable Use Policy (AUP) that guests must accept at the captive portal. The technical filtering policy must mirror the AUP. At a minimum, block: known malware and phishing domains; CSAM (integrate with databases such as the Internet Watch Foundation blocklist); P2P file-sharing protocols; and adult content for family-appropriate venues.

Step 2: Configure the Captive Portal and Authentication

Ensure the captive portal mandates authentication. Anonymous access is the enemy of the audit trail. Implement session limits and ensure DHCP lease times are optimised for high-turnover environments. For Hospitality deployments, integrate with the Property Management System (PMS) to authenticate guests against their booking reference.

Step 3: Deploy DNS Filtering and Gateway Rules

Integrate a cloud DNS filtering service. Configure the network gateway to intercept all outbound DNS requests on port 53 and force them through the approved filtering service. Implement firewall rules to block known DoH endpoints. Configure application-layer rules to drop P2P protocol traffic.

Step 4: Whitelist Critical Services

Ensure critical venue services are whitelisted before go-live. If your venue uses location services or navigation tools — for example, Purple Launches Offline Maps Mode for Seamless, Secure Navigation to WiFi Hotspots — ensure the relevant endpoints are accessible. Also prepare support teams for common post-deployment issues; filtering can occasionally cause connectivity anomalies, as discussed in Solving the Connected but No Internet Error on Guest WiFi .

Step 5: Test and Validate

Before going live, conduct a structured test: attempt to access known blocked categories from a guest device, verify the block page is displayed, verify the audit log captures the session, and confirm legitimate traffic is unaffected.

Best Practices

Dynamic Threat Intelligence: Static blocklists are obsolete within hours of publication. Ensure your filtering engine uses real-time, continuously updated threat intelligence to categorize new domains as they emerge. Threat actors register new domains daily specifically to evade static lists.

Granular Policy Control: Avoid blanket bans that disrupt legitimate business. Blocking all video streaming may be appropriate for a corporate office network but would be entirely inappropriate for a hotel. Define policies per SSID, per venue type, or per time of day where the platform supports it.

Encrypted Traffic Management: As TLS 1.3 and DoH become standard, relying solely on DNS is insufficient. Evaluate hardware capable of Server Name Indication (SNI) inspection as a middle ground between full DPI and DNS-only filtering. SNI inspection reads the unencrypted server name in the TLS handshake without decrypting the payload, offering category-level blocking with minimal throughput impact.

Compliance Logging: Maintain connection logs — MAC address, assigned IP, timestamp, authenticated identity — in compliance with local data retention laws. Under GDPR, do not log full browsing history; log only connection metadata. Ensure logs are encrypted at rest and access-controlled.

Troubleshooting & Risk Mitigation

Common Failure Modes

The DoH Bypass: Guests using modern browsers configured to use DNS over HTTPS will bypass standard DNS filters. Mitigation: Maintain an updated blocklist of DoH provider IPs at the firewall level and redirect all port 53 traffic via NAT.

MAC Randomization: Modern iOS and Android devices randomize MAC addresses per SSID, breaking traditional device tracking. Mitigation: Rely on session-based authentication tied to the captive portal login, rather than persistent MAC tracking. The session ID, not the MAC, becomes the audit key.

Over-Filtering and False Positives: Aggressive filtering blocks legitimate traffic, generating helpdesk tickets and degrading the guest experience. Mitigation: Implement a rapid whitelist review process. Monitor blocked domain logs weekly and whitelist confirmed false positives within 24 hours.

Policy Drift Across Sites: In multi-site deployments, manually managed policies diverge over time. Site A may have an outdated blocklist while Site B is current. Mitigation: Enforce centralised, cloud-managed policy distribution with version control. All sites must pull from the same policy baseline.

ROI & Business Impact

The Return on Investment (ROI) for content filtering is primarily measured in risk avoidance. A single copyright infringement lawsuit or ICO enforcement action can cost tens of thousands of pounds — far exceeding the annual cost of a filtering solution. The table below illustrates the cost differential:

Furthermore, filtering improves overall network performance. By blocking bandwidth-heavy P2P traffic and malware botnets, you preserve throughput for legitimate guests, improving the user experience and reducing infrastructure strain. When combined with a robust WiFi Analytics platform, the network transforms from an unmanaged liability into a secure, data-generating asset that drives measurable business outcomes.