The digital advertising landscape is unfortunately rife with all manner of attack ads that try to claim that a user has been infected with a virus and must immediately install a virus protection software from the site. Other attack ads are also quite commonplace, and they all use social engineering to make themselves seem more legitimate than might have been the case otherwise.
With all of that having been said and now out of the way, it is important to note that researchers at Georgia Tech may have finally created a solution for this problem. They have developed an add-on called Titan which apparently has the ability to block 100% of all attack ads and deceptive ads that you might see on a search engine results page.
The main issue with the advertising industry is that it can be a great place for malicious actors to find unsuspecting victims. This new tool is important because of the fact that this is the sort of thing that could potentially end up keeping these ads away from users, thereby preventing them from falling prey to them anytime soon.
The team that developed this new tool was spearheaded by Zheng Yang, and they collectively created a dataset that comprised over 100,000 sites from around the world. This revealed nearly 1,500 attacks, or 1,479 to be precise. Tech support, software downloads and scareware are some of the most common things being offered in these attack ads. Dating and notification scams also pop up quite frequently, as do ads that claim that you have won some kind of a prize.
One of the most astounding facts from this study is that the tool had false positive rate of just 2.57%. This means that over 97% of all of the ads that were blocked did turn out to be attack ads to one extent or another. As a result, legitimate ads have a very low level of likelihood of getting blocked, although it will be interesting to see what the results look like in a real world setting when it is tested.
Read next: A Recent Study Sheds Light on AI Hallucinations and What Society Thinks of Them
by Zia Muhammad via Digital Information World
No comments:
Post a Comment