๊ฐ€์ž… ํ›„ ์ดˆ๋Œ€ ๋งํฌ๋ฅผ ๊ณต์œ ํ•˜๋ฉด ๋™์˜์ƒ ์žฌ์ƒ ๋ฐ ์ดˆ๋Œ€ ๋ณด์ƒ์„ ๋ฐ›์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

Maxime Rivest ๐Ÿง™โ€โ™‚๏ธ๐Ÿฆ™๐Ÿง
@MaximeRivest
Searching for the wise way to interact with AI. Experiments are on GitHub Chat bot are likely not it. We probably need more AI commands and einks, less chats
๊ฐ€์ž… January 2018
966 ํŒ”๋กœ์ž‰ ์ค‘    10.6K ํŒฌ
Coding agents feels very human. So we naturally get foul in using then like humans (e.g. building teams and having concervations). This will prevent the most important breakthroughs and use cases. They are only deep neural networks. Neural nets have very different sets of characteristics... and tradeoffs! then human do. We dont know much about using these tools, we know a lot about delegating and teaming up with humans. This difference explains why what seems to work feels very human and we don't see many 'systems'/programs/automations. I propose that it's simply a matter of time, because we need to figure it out, not because it's an inferior approach.
๋” ๋ณด๊ธฐ