Transcript · English → Chinese
Discussion on Simulation Theory and AI
我们生活在模拟世界里吗？
是的。
我们是吗？
是的。
你为什么这么认为？
从统计学上看，如果我们正走在创造
智能体的道路上，而我们看来很快就 能做到，
并且我们在虚拟世界方面取得了相 当大的进展。
把这两点结合起来，就意味着任何人 都能创造
全新的
文明，充满智能生命的世界。
就像给孩子玩的电子游戏。 每个孩子都会玩几款游戏。
所以就会有数十亿个那样的虚拟
世界，里面住着和你我一样的生命。
你是，呃，虚拟世界中的一员，还是
唯一真实世界中的一员，概率有多大 ？
如果我们生活在模拟世界里， 谁在运行它？
模拟者。
那些模拟者是谁？
我毫无头绪。我就在模拟之中。 我们得逃出去。
我们需要逃离， 并从外界获取真正的知识。
你可以通过观察这项 设计，了解设计者的一些情况。
所以比如说， 你知道他们不在乎人类
的苦难。
如果我们生活在模拟中， 你是真实的人吗？
你只是某人写出的角色吗？
所以，“真实”这个概念，呃， 非常取决于具体领域。
人们说，你永远不会被
模拟的雨淋湿，
但那不是真的。如果你 是模拟中的一个角色，
你确实
会在 那个模拟中被模拟的雨淋湿。
那么我在这个世界是真实的吗？ 我会感到痛苦吗？我会感到快乐吗？
我会坠入爱河吗？是的，我是真实的， 就那种意义而言。
我是否
以同一个化身存在于 真实世界？我不知道。
你相信我们生活在模拟中 有多久了？
五年。
那是你读到的、学到的， 还是跟别人聊来的？
尼克·博斯特罗姆关于模拟的论文 对许多
人很有影响力，不过，呃， 我认为我们可以比
他所提出的更深入地探讨。 我认为他有大约百分之二三十的
把握。 我认为我们能达到更高的确信程度。
但如果我们生活在模拟之中， 最危险的
实体是人工超级智能， 还是运行
这个模拟的人？
运行它的人，迄今为止， 姑且算
一百五十亿个感知中的年头
都没有毁灭我们。
所以如果他们有计划， 他们还没有实施。
他们本有机会， 而且拥有完全的控制权。
他们随时都能关闭它， 所以我们不必担心这是一项
新的发展。 而我们将其带入存在
潜在的竞争性超级智能 可能会触发关机。
你觉得我们会被关掉吗？
如果这是个模拟世界。
那会是什么样子？
熄灯。
熄灯。
熄灯。
应该把灯关掉。
熄灯。
好。你也研究人工智能意识
是的。
你觉得
自己有意识吗？
我知道自己有意识。
你怎么知道？
因为我能直接接触到自己的 内部体验。
好。
这是我唯一确定的事。 我不知道这个世界是不是真的。
你不知道我是不是真的。
我不知道你是否有意识， 但我知道自己有内在的
感知。
好。
我体验，故我有意识。
所以你就是这样验证自己的 意识吗？
这是主观的。 意识的难题
在于内在状态。
当一只蝙蝠是什么感觉？当一名 罗马人又是什么感觉？
有人能验证
人工智能的意识吗？
所以我有一篇论文， 人们要么讨厌，要么喜欢，
提出了一种测试某些内在状态的方 法。
它无法明确检测出所有可能的 意识状态，
但如果你的状态和我的相似， 我大致能判断你
是否拥有它们，因为你在关于幻觉的 多项选择题中
给出了相同的回答。
如果一个超级智能人工智能
能在你行动前预测你的下一步，
那么你是否还拥有
自由意志呢？
这取决于你如何定义自由意志。 它关乎不可预测性吗？
史蒂芬·沃尔夫勒姆对元胞 自动机有出色的研究，而我们知道
其中一些无法被归约为
从……角度而言
运行这项计算。 你必须运行它，才能得知
那个自动机
的状态，但它完全是确定性的。
不会有意外，也没有随机性。
它有自由意志吗？ 你无法预测它会做什么，
但如果你有一台更快的 硬件设备，就能精确预测它会做什么 。
嗯。如果我们身处模拟中， 你曾做出的每个选择，
嗯。如果我们身处模拟中， 你曾做出的每个选择，
你写过的每篇论文，
你发出过的每次警告，
那是否只是模拟按其 既定轨迹运行？
按定义来说，是的。
End of transcript · 815 words · machine translated
Original transcript (English)
Are we living in a simulation? Yes. We are? Yes. Why do you think that?
Statistically seems more likely if we are on a path to create intelligent agents, which it looks like we will soon, and we made pretty good progress in virtual worlds. Just putting those two together means anyone can create new
civilizations, worlds populated by intelligent beings. Let's say it's like video games for kids. Every kid plays a couple games. So you have billions of those virtual worlds populated by beings just like me and you. What are the chances you are one of, uh, virtual ones versus the
only real one? If we're living in a simulation, who's running it? Simulators. And who are those simulators? I have no idea. I'm inside the simulation. We need to escape.
We need to escape and gain real knowledge from outside. You can learn certain things about the designer by looking at the design. So for example, you know they don't care about human suffering. If we're living in a simulation, are you a real person?
Are you just a character someone wrote? So real is, uh, very domain specific terms. People talk about you can never get wet from simulated rain, but that's not true. If you are a character in a simulation,
you literally will get wet from a simulated rain in that simulation. So am I real in this world? Do I feel pain? Do I feel pleasure? Can I fall in love? Yes, I'm real in that sense. Am I
represented with the same avatar in the true world? I don't know. How long have you believed that we're living in a simulation? Five years. Is that something you read, you learned, you talked to somebody? Um, Nick Bostrom's paper on simulation was very influential for many
people, but, uh, I think we can take it even deeper beyond what he suggested. I think he's about twenty, thirty percent confident. I think we can get higher levels. But if we're living in a simulation, is the most dangerous entity artificial super intelligence, or is it whoever's running
the simulation? Whoever's running it so far for, let's say, fifteen billion of perceived years was not destroying us. So if they have a plan, they haven't done it yet.
They had a chance, and they have complete control. They can shut it off at any point, so we don't have to worry about this as a new development. Whereas us bringing into existence potentially a competing super intelligence may trigger shutdown. You think we can be turned off?
If it's a simulation. And what does that look like? Lights out. Lights out. Lights out.
Should turn the light off. Lights out. Okay. You study AI consciousness As well, yes. Do you think that
you are conscious? I know I am conscious. How would you know? Because I have direct access to my internal experiences. Okay.
That's the only thing I know for sure. I don't know if this world is real. You don't know if I'm real. I don't know if you are conscious, but I know that I have internal perceptions. Okay.
I experience, therefore, I'm conscious. So that's how you verify your own consciousness? It is subjective. The hard problem of consciousness is about internal states. What is it like to be a bat? What is it like to be a Roman?
Is anyone gonna be able to verify an AI's consciousness? So I have a paper which people either hate or love, proposing a test for some internal states. It's not definitive where it detects all possible states of consciousness,
but if yours are similar to mine, I can kind of tell that you have them because you answer the same way in multiple choice questions about illusions. If a super intelligent AI can predict your next move before you make it,
then do you ever have free will to begin with? It depends on how you define free will. Is it about unpredictability? So Stephen Wolfram has excellent research on cellular automata, and we know that some of them cannot be reduced in
terms of running the computation. You have to run it to find out the state of that automata, but it's fully deterministic. There is no surprises, no randomness.
Does it have free will? You cannot predict what it's going to do, but yet you can precisely predict what it's going to do if you have a faster hardware device. Hmm. If we're in a simulation, every choice you've ever made, every paper you've written,
every warning you've given, is that just the simulation running its course? By definition.
