Between 1 and 3 November 2026, readers of VI.nl, GeenStijl and other Dutch sites could be exposed to iPhone exploit code without clicking anything. The code did not come from the publishers. According to Privacy Affairs, someone broke into the system that manages the sites' ads and changed an ad tag. NOS named the ad company as NextDay, the advertising arm of publisher PXR. NextDay says it serves 150 sites. That does not mean all 150 ran the code, and nobody has said how many people saw it.
What happened, and what is still unknown
Privacy Affairs reports that the malicious script collected data on each visit and tried to attack iPhones that had missed security updates. Its timeline, in Dutch time, shows the code live from about 09:50 on 1 November to 10:07 on 3 November. A second period ran on 3 November from about 12:06 to 17:20. A third, short change ran on 11 November from about 15:34 to 15:50, and the sites do not yet know what it did. NOS reported on 15 November that at least 14 Dutch sites and one Belgian site had warned readers. The warnings came on the evening of 14 November, roughly two weeks after the exposure began.
No notice names the exploit kit. GeenStijl linked it to Coruna, but Privacy Affairs notes that Coruna only works up to iOS 17.2.1, so it cannot explain cases on iOS 18. Privacy Affairs offers DarkSword as a better match for the iOS 18 cut-off and labels that its own reading, not a confirmed finding. This analysis treats the kit's identity as unconfirmed.
Google's Threat Intelligence Group documented Coruna in March 2026. It has five iOS exploit chains and 23 exploits, covering iOS 13.0 through 17.2.1. Google says it cannot attack the latest iOS versions. Its documented delivery was compromised websites and fake sites, not advertising. So the exposure window here is limited to devices that skipped updates, and the fix for those users is simple: update.
The case for tougher rules
The strongest argument for new regulation is structural. A publisher cannot audit every script an ad stack loads on each page view. Readers cannot judge a site's ad supply chain at all. Publishers earn the revenue, but the harm lands on readers and the supplier is invisible to them. Mandatory disclosure and tighter liability for ad intermediaries would put the cost on the party best placed to prevent the harm.
Why existing law already covers most of this
The Netherlands has just moved on exactly this kind of supply-chain risk. The Cyberbeveiligingswet (Cbw) implements the EU NIS2 directive. The Tweede Kamer adopted it on 15 April 2026, and the Eerste Kamer adopted it on 7 July 2026. The government describes its core obligations as a duty of care (zorgplicht), a reporting duty (meldplicht) and a registration duty (registratieplicht). I could not confirm from the sources I fetched that the law has been promulgated or is in force.
Whether a regional ad network or a small publisher falls within the Cbw's scope is a question I have not been able to answer from the sources reviewed. NIS2 scoping turns on sector and size, and many independent publishers will be outside it. That is a real gap, but the answer is not a new ad-tech statute. It is for regulators and industry to use existing duties and contracts to push security expectations down the supply chain.
The GDPR already supplies a reporting trigger. Privacy Affairs notes that visit data such as IP addresses is personal data, and that a breach must be reported to the data protection authority within 72 hours unless it is unlikely to put people at risk. The Autoriteit Persoonsgegevens (AP) sets the same 72-hour rule from discovery and says late reports need justification. It also says a processor can report on the controller's behalf only if that has been agreed in writing. That matters here, because the ad network is the likely processor and each publisher is a controller. None of the publisher notices says whether the AP was notified, and Privacy Affairs has asked PXR and NextDay.
That is where accountability should be tested. The questions are who discovered the compromise, when, who told the publishers, and why readers heard only on 14 November. A two-week gap between the start of exposure and a public warning is the part that regulators and publishers can scrutinise under rules already on the books.
A proportionate response
The pro-innovation position is not that nothing should change. It is that the response should match the evidence, and the evidence so far points to an access-control and detection failure at one ad-management system. It does not point to a flaw in open, ad-funded publishing. Banning programmatic advertising or imposing strict liability on every ad intermediary would hit small publishers hardest. Many of them depend on shared ad stacks, as the 150 sites on NextDay do, and cannot build their own.
Three measures follow from this incident without new legislation:
- Contract terms. Publishers should require ad suppliers to use multi-factor authentication on tag-management consoles, keep change logs, and notify within a fixed time after discovery.
- Technical controls. Subresource integrity, content security policies and monitoring of third-party script changes would shrink the window in which a swapped tag can run.
- Transparent post-incident reporting. PXR and NextDay should say publicly how the intruder got in, what the three time windows did, and whether the AP was notified.
Users have a part too, though the sites' own notices correctly put the burden on suppliers. By Privacy Affairs' account, iPhone XS, XR and newer models running iOS 18.7.2 or older are at risk. So are older models such as the iPhone 6s, 7, 8, X and first-generation SE if they lack the March 2026 update. Updating closes the hole.
The test of the Cbw and the GDPR is whether they change behaviour in cases like this one. If the facts show a late or missing breach notification, the AP has the tools to act. If they show a small publisher that did what it reasonably could, the law should not punish it for a supplier's failure.