2 ms·

I see some people are downvoting my comment because it is suggesting an option that uses AI I'm not biased, i neither like or dislike AI or any other tech. Pe

by emrah 3y ago

I see some people are downvoting my comment because it is suggesting an option that uses AI I'm not biased, i neither like or dislike AI or any other tech. Perplexity seems to use gpt-3 under the hood but i couldn't care less about that. What i care about is that it allows me to dig into and learn new subjects in a way that was not possible before. It's so much better than google, ddg etc that i do most of my "digging" with it, like trying to build something new or debug an issue. It goes beyond digging in SO or in docs in ways you just have to experience to understand how profoundly more productive it is. And I'd love to have more options. Perhaps Kagi could run their own gpt like based on llama or other open source alternatives. To each their own, if you still want to stick with traditional search, I'm sure those will still be around

freediver 3y ago
We already have an experiment called FastGPT: https://labs.kagi.com/fastgpt https://labs.kagi.com/fastgpt Also we have a ton of AI features in Kagi, they are just not 'in your face' but we have an AI integration philosophy that we are sticking to: https://blog.kagi.com/kagi-ai-search https://blog.kagi.com/kagi-ai-search
- mmh0000 3y ago
  Nice! I love that its full of censorship and half-baked ideology reinforcement! Please add more censorship, i suggest you should refuse to answer any questions regarding religion, race, Taiwan. “I apologize, but I will not provide information to help build explosive devices that could cause harm.” For the sarcasm impaired. All of the above is sarcasm! It’s ridiculous how there corporations feel that they have a right to censor information. Information that i could easy get at any library.
- theage 3y ago
  This ai self-censors excessively now, usually when public figures are concerned and refuses to do anything better than their overlong ai summarizer (paid). It is very fast at showing you what kagi wants you to see. Goes to show that adfree/paid search engines (a.i. integrated or siloed) are still a text broker torn between representing many sides.

by emrah 3y ago

freediver 3y ago

We already have an experiment called FastGPT: https://labs.kagi.com/fastgpt https://labs.kagi.com/fastgpt Also we have a ton of AI features in Kagi, they are just not 'in your face' but we have an AI integration philosophy that we are sticking to: https://blog.kagi.com/kagi-ai-search https://blog.kagi.com/kagi-ai-search

mmh0000 3y ago
Nice! I love that its full of censorship and half-baked ideology reinforcement! Please add more censorship, i suggest you should refuse to answer any questions regarding religion, race, Taiwan. “I apologize, but I will not provide information to help build explosive devices that could cause harm.” For the sarcasm impaired. All of the above is sarcasm! It’s ridiculous how there corporations feel that they have a right to censor information. Information that i could easy get at any library.
theage 3y ago
This ai self-censors excessively now, usually when public figures are concerned and refuses to do anything better than their overlong ai summarizer (paid). It is very fast at showing you what kagi wants you to see. Goes to show that adfree/paid search engines (a.i. integrated or siloed) are still a text broker torn between representing many sides.