Lex Fridman interviews Nick Bostrom.

Nick Bostrom is a philosopher at University of Oxford and the director of the Future of Humanity Institute. He has worked on fascinating and important ideas in existential risks, simulation hypothesis, human enhancement ethics, and the risks of superintelligent AI systems, including in his book Superintelligence. I can see talking to Nick multiple times on this podcast, many hours each time, but we have to start somewhere. This conversation is part of the Artificial Intelligence podcast.

Time Index:

  • 0:00 – Introduction
  • 2:48 – Simulation hypothesis and simulation argument
  • 12:17 – Technologically mature civilizations
  • 15:30 – Case 1: if something kills all possible civilizations
  • 19:08 – Case 2: if we lose interest in creating simulations
  • 22:03 – Consciousness
  • 26:27 – Immersive worlds
  • 28:50 – Experience machine
  • 41:10 – Intelligence and consciousness
  • 48:58 – Weighing probabilities of the simulation argument
  • 1:01:43 – Elaborating on Joe Rogan conversation
  • 1:05:53 – Doomsday argument and anthropic reasoning
  • 1:23:02 – Elon Musk
  • 1:25:26 – What’s outside the simulation?
  • 1:29:52 – Superintelligence
  • 1:47:27 – AGI utopia
  • 1:52:41 – Meaning of life

Lex Fridman interviews Grant Sanderson is a math educator and creator of 3Blue1Brown, a popular YouTube channel that uses programmatically-animated visualizations to explain concepts in linear algebra, calculus, and other fields of mathematics.

OUTLINE:

0:00 – Introduction
1:56 – What kind of math would aliens have?
3:48 – Euler’s identity and the least favorite piece of notation
10:31 – Is math discovered or invented?
14:30 – Difference between physics and math
17:24 – Why is reality compressible into simple equations?
21:44 – Are we living in a simulation?
26:27 – Infinity and abstractions
35:48 – Most beautiful idea in mathematics
41:32 – Favorite video to create
45:04 – Video creation process
50:04 – Euler identity
51:47 – Mortality and meaning
55:16 – How do you know when a video is done?
56:18 – What is the best way to learn math for beginners?
59:17 – Happy moment

Microsoft Research marks its 100th episode with with Gurdeep Pall and Dr. Ashish Kapoor talking about autonomous systems.

There’s a lot of excitement around self-driving cars, delivery drones, and other intelligent, autonomous systems, but before they can be deployed at scale, they need to be both reliable and safe. That’s why Gurdeep Pall, CVP of Business AI at Microsoft, and Dr. Ashish Kapoor, who leads research in Aerial Informatics and Robotics, are using a simulated environment called AirSim to reduce the time, cost and risk of the testing necessary to get autonomous agents ready for the open world. 

See more on this and other Microsoft Research podcast episodes:
https://www.microsoft.com/en-us/research/blog/category/podcast/

Samuel Arzt shows off a project where an AI learns to park a car in a parking lot in a 3D physics simulation.

The simulation was implemented using Unity’s ML-Agents framework (https://unity3d.com/machine-learning).

From the video description:

The AI consists of a deep Neural Network with 3 hidden layers of 128 neurons each. It is trained with the Proximal Policy Optimization (PPO) algorithm, which is a Reinforcement Learning approach.