1. It uses “perceptual ad blocking”
Most ad blockers in use today look for the footprints of digital ads. For example, they scan the contents of a page, identifying snippets of code and URLs that are commonly associated with ads and ad networks. This approach is very effective at blocking ads served by major ad players like Google, but it’s far less effective at weeding out ads that are served by publishers themselves, including native ads.
The university researchers’ ad blocker uses computer vision technology to analyze the contents of a web page much the same way a human would. As Vice’s Jason Koebler explained…
…it uses optical character recognition, design techniques, and container searches (the boxes that ads are commonly put in on a page) to detect words like “sponsored” or “close ad” that are required to appear on every ad, which is what allows it to detect and block Facebook ads.
2. It could be 100% effective at blocking ads
According to a paper published by the Princeton and Stanford researchers who created the new ad blocker, so long as advertisers and publishers adhere to the disclosure standards promulgated by regulatory agencies like the Federal Trade Commission (FTC), “a perceptual ad blocker will have a 100% recall at identifying ads governed by that standard.”
3. The ad blocker is difficult if not impossible to detect
The new computer vision-based approach allows the Princeton and Stanford ad blocker to more stealthily block ads by taking advantage of techniques normally employed by malware. In fact, when tested on 50 websites employing anti ad-blocking scripts, the university’s ad blocker was able to block ads without being detected 100% of the time.
Because it stealthily blocks ads, the super ad blocker is able to avoid detection by anti ad-blocking techniques that publishers commonly use to thwart users who browse their sites with ad blockers.
That’s obviously not going to be good news for the growing number of publishers that are using anti ad-blockers to cut off access to people using the technology, and could encourage some of them to erect more restrictive paywalls.
4. The technology is currently available in a proof-of-concept that doesn’t block ads
While it’s probably only a matter of time before perceptual ad blocking technology makes its full debut, for now, the Princeton and Stanford researchers decided to release a limited proof-of-concept in the form of a Chrome extension that doesn’t actually block ads. Instead, the extension highlights the ads it identifies.
5. Perceptual ad blocking isn’t the only new approach to ad blocking that could have a big impact
While it would appear that perceptual ad blocking has the potential to end the ad blocking wars, handing victory to consumers and defeat to publishers and advertisers, this new approach to ad blocking isn’t the only one that publishers and advertisers need to worry about.
In a paper detailing their perceptual ad blocking tech, the Princeton and Stanford researchers also presented another technique that could make ad blocking more effective. Under this approach, a browser extension would create two copies of a page, blocking ads in the one displayed to users. The end result would be that publishers would again have no way to identify users who are blocking ads.