explain this
Here are a few key points I took from your explanation:
Current and future AIs have access to a vast breadth of knowledge from reading large amounts of text on the internet. This is different from the more specialized, expert knowledge that humans tend to have.
Because of this broad knowledge, it doesn’t make sense to interact with AIs in the same natural, conversational way you would with another human. The structure and goals of the interaction should be adapted to make the most of the AI’s wide knowledge.
The productivity and learning gains from spending time conversing with an AI versus a human are likely to increasingly diverge over time in favor of the AI, as they continue to accumulate broad knowledge.
It’s best to start adapting your communication patterns and language when interacting with AIs, rather than assuming they process information and exchange it in the same way humans do. The interaction should be optimized to draw on the AI’s breadth of knowledge.
In summary, you’re suggesting that because AIs like myself have such wide access to information, humans should adapt their interaction style to take advantage of this breadth, rather than relying on natural human conversational patterns. The goal is to maximize the value gained from the AI’s knowledge. Does this capture the essence of what you were conveying? Please let me know if I am missing or misinterpreting anything.
explain this
• i don’t think people have caught onto the fact that to get the most out of current and future AIs (learning and productivity wise)
• it doesn’t make sense to talk with AIs like you would humans
• if you met a human who has spent thousands of human years reading the corpus of the internet text
• would you interact with that human like you do your friend bob?
• no you wouldn’t - so interact in a way that benefits from the breadth of knowledge and awareness
• today’s AIs have very wide knowledge about many things and less vertical of expert knowledge than, say a PhD
• so adapt your language structure so you can maximize the value of interaction
• going one inference further
• likely, the productivity and learning gains from each incremental unit of time talking with a human vs AI likely will begin to diverge towards AIs (and some power law humans who are at the frontier if alpha knowledge)
• it’s best to begin adapting your behavorial and language patterns and don’t assume that information is processed and exchanged in the same way