Главная
Study mode:
on
1
Introduction
2
What is AI
3
What if we succeed
4
Benefits of AI
5
Limitations of AI
6
Nuclear energy
7
AI systems
8
Algorithms in social media
9
Standard model
10
Principles
11
Example
12
Research questions
13
Real humans
14
Tradeoffs
15
Avengers
16
The social aggregation theorem
17
Open source game theory
18
Avoiding failure mode
19
Nontechnical version
20
Summary
21
QA Can AI perform better than humans
22
AI and preferences
23
AI and happiness
24
AI in other disciplines
Description:
Explore the future of artificial intelligence and its potential impact on humanity in this thought-provoking 58-minute lecture by Stuart Russell from the University of California. Delve into the possibilities of AI surpassing human capabilities in decision-making scenarios and the implications this may have on our future. Examine both the potential benefits and concerns surrounding AI development, including the need for a fundamental reorientation of the field. Learn about the importance of building AI systems that are truly beneficial for humans, and discover the new research avenues this approach opens up. Investigate key topics such as AI capabilities, limitations, ethical considerations, and the principles needed for creating human-compatible AI. Engage with concepts like the standard model of AI, social media algorithms, and open-source game theory. Conclude with a Q&A session addressing AI performance, preferences, happiness, and its applications in various disciplines.

Human-Compatible Artificial Intelligence - Stuart Russell, University of California

Alan Turing Institute
Add to list