We need AI that detects clickbait titles. The browser could open links in the background, scan the articles and see (using AI) if they correspond to the title of the link.
If not, then the link becomes red (or something).
Alternatively, a crowd-sourced solution could also help.
First it was not OK to have click-bait titles, as readers felt tricked. If you got accused of click-bait, you tried adjusting the title so people didn't feel tricked.
Then people starting to do as click-baity titles as they could and get away with, until they were found out, then they adjusted until the next blog post.
Now people write click-baity titles and even admit it in the first paragraph! And still, judging by the amount of blogposts that does this, people read through the post, even if they got bait-and-switched.
Same sentiment around selfies. At first it was frowned upon, what kind of narcissist person takes photos of themselves?! Then it was OK for the most superficial people to do it, they are narcissists after all. But suddenly the tide turned around and now it's weird if you're not posting selfies every now and then.
The dish-o-tron series is intended to be a fun (ironic) deeplearning tutorial. The title for the first part "dish-o-tron – no more dirty dishes tanks to AI" is in a sense correct because the final goal is to prevent mysteriously appearing dirty dishes in the community kitchen sink.
However, we are fully aware that it also could promise too much at first glance. Hence, we thought it would be witty to have this kind of disclaimer. As always, humor is a matter of taste so sorry if you feel mislead.
