RCreddit.com·
Not on the current live radar
Eliezer Yudkowsky vs Cal Newport: Discuss
A discussion on AI safety highlights differing perspectives from Cal Newport and Eliezer Yudkowsky. Newport suggests that dangers arise when LLMs are integrated into autonomous loops with tools, memory, and feedback, potentially leading to uncontrolled behavior even without inherent AI goals. Yudkowsky, conversely, argues that a sufficiently capable system, regardless of consciousness or human-like intentions, can become a dangerous optimizer that is difficult to control.
Time & source
- Ingested
- 09/09, 07:00 UTC+0
- Source type
- Dev community
Discussion trend
No comparison yet
The percentage is based on collected discussion signal, not new comments or independent people. The curve only compares the same topic across time.
Article
Full text isn't available here.
Read at source →