Until _that_ company gets overrun by MBAs who are profit-driven then they start injecting ads into the results.
It will come in the vein of "we are personalizing the output and improving responses by linking you with vendors that will solve your problems".
Found companies with people that share your values. Hire people that share your values. Reject the vampires. Build things for people.
Competing with bad actors is very, very hard. They will be fat with investor money, they will give their services away, and commonly they are not afraid to do things like DDOS to raise your costs of operations.
What if the user is a conservative voter and considers anything counterpoint to their world view the worst part of the internet and removes all instances of it from their daily lives? Not to say that isn’t already happening but they are consciously making the choice, not some AI bot. I can see something like this making the country even more polarized.
Growing up as a southern evangelical before the internet, I can promise you that there has never been a modern world without filter bubbles.
The concept of "fake news" is not new, either. There has been general distrust of opposing ideas and institutions for as long as I've been alive.
And there's an entire publishing and media ecosystem for every single ideology you can imagine: 700 Club, Abeka, etc. Again, this all predates the internet. It's not going anywhere.
The danger isn't strictly censorship or filter bubbles. It's not having a choice or control over your own destiny. These decisions need to be first class and conscious.
Also, a sure fire way to rile up the "other team" is to say you're going to soften, limit, or block their world view. The brain has so many defenses against this. It's not the way to change minds.
If you want to win people over, you have to do the hard, almost individual work, of respecting them and sharing how you feel. That's a hard, uphill battle because you're attempting to create a new slope in a steep gradient to get them to see your perspective. Angering, making fun, or disrespecting is just flying headfirst into that mountain. It might make you feel good, but it undoes any progress anyone else has made.
A sufficiently advanced personal assistant AI would use multimodal capabilities to classify spam in all of its forms:
- Marketing emails
- YouTube sponsorship clips
- Banner ads
- Google search ads
- Actual human salespeople
- ...
It would identify and remove all instances of this from our daily lives.
Furthermore, we could probably use it to remove most of the worst parts of the internet too:
- Clickbait
- Trolling
- Rage content
I'm actually really looking forward to this. As long as we can get this agent into all of the panes of glass (Google will fight to prevent this), we will win. We just need it to sit between us and everything else.