More than half of all internet users have installed an ad-blocker; Doc Searls calls it the largest consumer boycott in history. As large as that boycott is, everyone – yes, everyone – should block every "programmatic" ad, because the entire system is so insecure that failing to do so could result in the total compromise of any or all of your digital devices.
That's the inescapable conclusion of Citizen Lab's latest blockbuster investigative report, "Exposing AdFection," which reveals how the largest ad-tech companies in the world, including Google and Facebook, allowed themselves to be hijacked by a "mercenary spyware" company that used the ad system to locate targeted individuals and serve "zero-click malware" that took over their computers and phones without them having to take any action:
https://citizenlab.ca/research/how-discover-tech-turned-targeted-ads-into-a-pipeline-for-spyware/
These tech giants amassed detailed surveillance dossiers on virtually every internet user, and they've assigned each of us unique, persistent identifiers, like a Social Security Number for spying. These companies also allow advertisers – or anyone pretending to be an advertiser – to serve blobs of code (which is all an ad is) to anyone based on those identifiers.
This allowed an Israeli spyware company called Discover-Tech (which masquerades as an ad-tech company) to sell totalitarian governments and sleazy corporate mercenaries the ability to pluck individual internet users out of the billions who log in every day and then serve "malicious ads" (that is, blobs of arbitrary code) to those targeted users. These malicious ads contained "no-click, zero-day" attacks, which meant that they exploited previously unknown vulnerabilities to take over devices without any user interaction. The mere act of loading a web-page or opening an app that relied on one of these gigantic ad-tech companies to serve ads was enough to compromise your device. No clicks needed, and no warning.
Among the targets of this attack campaign were the computers of workers at Al Jazeera. They were attacked with malware from the mercenary spyware company Candiru (which now calls itself "Integrity Labs" because irony is not dead). Beyond Candiru, Discover-Tech also facilitates attacks by numerous criminal mercenary spyware companies, serving ads on hundreds of thousands of sites and apps.
Discover-Tech was "certified" as a safe supplier by Google, Yandex, Microsoft, Beeswax (a division of Comcast) and IAB, the ad-tech industry's leading professional association. Ironically, Discover-Tech even attacked users by serving them malicious GDPR notices that appeared to let users opt out of commercial surveillance.
Because Discover-Tech was certified by all these industry bodies and firms, it had access to more than four billion ad-auctions, meaning that on billions of occasions, it was able to collect the identities of users loading apps and visiting websites, check them against the list of targets, and attack any victims it discovered through this dragnet.
Discover-Tech is an Israeli company with ties to other Israeli spyware and hacking companies, including through its board, whose members included Eric Banoun ("One of the pioneers of offensive cyber in Israel" –Haaretz). They have overlaps, relationships and affiliations with a rogue's gallery of Israeli companies, including the notorious NSO Group, who routinely sell cyber-weapons to the world's worst governments and whose products have been used to attack journalists, embassy workers, and activists.
The whole incident highlights the monumental insecurity of the ad-tech industry, whose indiscriminate and voracious surveillance and manifestly insufficient vetting and security auditing now expose every internet user in the world to terrifying targeted attacks. These companies – including Google, a thrice-convicted monopolist whose surveillance tendrils are wrapped around the whole world – claim that they operate their networks with prudence and oversight. But the proof is in the pudding: for years, they've played host to a murderer's row of cyber-mercenaries who used their infrastructure to attack journalists and others.
The only way to avoid this is to block ads. This is easy enough to do in a browser (which is why the majority of internet users have installed an ad-blocker in their browsers), but it's illegal to reverse-engineer and modify an app so you can block the ads in it:
https://pluralistic.net/2026/07/21/dickovers/#block-element
Many websites run anti-ad-block scripts that push out nagging, guilt-tripping notices that call you a thief and a cheapskate for blocking ads. Ignore those notices: if a website is running "programmatic" ads that allow arbitrary third parties to spy on billions of people and target individuals for malicious device takeovers, they are charging too high a price to see their content.
Some sites go further: their notices demand that you either turn off your ad-blocker or pay for a subscription. This "pay or okay" scheme is illegal in Europe, California and other territories with strong privacy laws, and they are morally bankrupt besides. Privacy shouldn't be a luxury item and everyone deserves to be secure from cyber-attacks:
https://noyb.eu/en/noybs-pay-or-okay-report-how-companies-make-you-pay-privacy
This report also provides the most comprehensive rebuttal to a dangerous idea that refuses to die: that we should improve competition in the ad-tech sector by assigning every internet user a mandatory, government-issued advertising ID to "democratize" ad-targeting and followup (which the sector calls "attribution"):
https://pluralistic.net/2024/08/07/revealed-preferences/#extinguish-v-improve
Despite all this, none of it is an indictment of ads per se. The problem here isn't advertising, it's surveillance. The ad-tech sector has convinced advertisers that targeting ads to users rather than content will massively improve ad performance. The evidence for this is very poor: targeted ads improve clickthrough rates by less than 5% (relative to content-based ads that relate to the substance of the page you're looking at), and they're incredibly expensive (and very dangerous!) to serve.
On top of all that, surveillance-based ads shift power away from publishers (who know more about the materials they publish than any ad-tech company ever will) to spying ad-tech companies (who can spy on users in ways that no publisher could ever match). Whatever minor additional revenues surveillance advertising can wring out of the market are totally swamped by the cost of the ad-tech sector's own operations, which the ad-tech sector makes up for by grossly underinvesting in security and by colluding to cheat publishers and advertisers, charging more for ads and paying less to the websites that run them:
https://en.wikipedia.org/wiki/Jedi_Blue
If we want to improve competition in advertising services, the answer isn't to facilitate spying (which also allows the world's scummiest cyber-mercenaries to stage and serve targeted attacks). We should ban surveillance advertising altogether, restoring context-based ads to their role as the way that publishers make an honest living without selling out their audiences:
https://pluralistic.net/2023/05/31/context-ads/#class-formation