▲ | deadbabe 3 days ago | |||||||||||||
I imagine in the future someone will make an Agent-First OS that is entirely built from the ground up to be run by AI and runs off the assumption that there are no human users or that their usage is limited. That will be interesting, imagine all the things you could do differently, the design choices you could make. You lose a lot by accommodating human ergonomics. | ||||||||||||||
▲ | Waterluvian 3 days ago | parent | next [-] | |||||||||||||
What might you imagine being different in an “agent first OS” compared to a terminal only Linux distribution? | ||||||||||||||
| ||||||||||||||
▲ | pjerem 3 days ago | parent | prev [-] | |||||||||||||
For now, I still have hard time believing in unattended agents. Be it using a computer or generating programs. I mean that would sure be a nice demo, but it’s too probabilistic to give AI agents real tasks (and it seems that isn’t going to change anytime soon). It’s all fun and games until it implies spending money and/or taking responsibility. And be it in personal life or in businesses, money and responsibility are vital things. Sure you can ask LLMs to generate a minesweeper game with custom rules or ask it to summarize headlines from HN. Releasing a program an unattended agent generated to real clients that pay you or asking it to order a non refundable flight ticket is something else. However, I can see the point of an agent that uses my computer while I watch it. | ||||||||||||||
|