關於這一集
Despite our best efforts, sometimes what we say we want isn’t precisely what we mean. Nowhere is that felt more acutely than when we’re giving instructions to a machine, whether that’s coding or offering examples to machine learning systems. We unpack the alignment problem in this encore edition of our oral history project.
We Meet:
Brian Christian, former University of Oxford researcher, and author of The Alignment Problem
Credits:
This episode of SHIFT was produced by Jennifer Strong with help from Emma Cillekens. It was mixed by Garret Lang, with original music from him and Jacob Gorski. Art by Meg Marco.
英語
美國
逐字稿 🔗
SHIFT — The Alignment Problem (Brian Christian)
Machine-generated transcript; may contain minor errors. Timecodes are per segment.
===== SEGMENT A =====
[0:02] There's a reason some stories resonate with generation after generation.
[0:07] Like the tale of the Sorcerer's Apprentice, where Mickey Mouse, at least in the version
[0:11] that I know, casts a spell to help him get his chores done more easily.
[0:16] But his instructions are followed a little too closely, and things go quite wrong.
[0:…
Machine-generated transcript; may contain minor errors. Timecodes are per segment.
===== SEGMENT A =====
[0:02] There's a reason some stories resonate with generation after generation.
[0:07] Like the tale of the Sorcerer's Apprentice, where Mickey Mouse, at least in the version
[0:11] that I know, casts a spell to help him get his chores done more easily.
[0:16] But his instructions are followed a little too closely, and things go quite wrong.
[0:…
免責聲明: 本頁面嵌入的 Podcast 和封面圖片來自 Jennifer Strong，歸其所有者所有，與 Listen Notes, Inc. 無關聯，也未獲得 Listen Notes, Inc. 背書。
