What I like about it is that it’s trained on lots of different sources (including, but not limited to Google, and Bing search results). It then strips out the ads, SEO blog spam, and other nonsense and tries to return the most relevant info for my query. It is leagues better than pure Google. Also, it uses its own LLM unrelated to OpenAI.

A bit unfortunate that I got downvoted for having an opinion and sharing it.

I know Lemmy likes to hate on AI, but my default search engine is http://perplexity.ai and it’s great

I can’t stand the topics you mentioned either and I make a big effort to filter them out whenever possible! There’s a tool called https://siftrss.com/ which can be the middle man between you and your RSS feeds. It allows you to filter out any articles containing black listed keywords which is super handy.

My biggest frustration about modern RSS feeds is that the articles are often incomplete. Some clients are able to get the full feed anyway but it usually results in broken formatting.

Here are my favorite news sources:

Nestlé VPN

I laughed out loud at this suggestion. Love it. I can imagine this could lead to some confusion if a user blocks a common name, like John Smith… but let me give it some more thought!

This is a great suggestion and is pretty similar to what some of the other commenters also requested. I am thinking to support multiple modes:

• Blur The Rich (current prototype)
• Block The Rich (your request)
• Mock The Rich (replace the names of these billionaires with less flattering nicknames e.g. “Space Karen”, etc)