Go to https://ground.news/ordinary to avoid media bias, manipulative algorithms, and know where your news is coming from.
TIMESTAMPS
00:00 Sexy Freeman Podcast Intro
01:54 Chapter 1: First Contact
02:25 Frank and the Perceptrons
04:25 Eliza
05:25 AGI
08:30 Chapter 2: Beating the System
10:00 Predicting the Future
11:06 Rodney and his Robots
12:09 Deep Blue
13:10 The Singularity
14:36 The Nick Land Tangent
19:35 Chapter Three: Infohazzards
20:56 The PaperClip Maximiser
22:00 LessWrong
24:23 Roko’s Basilisk
27:19 Chapter 4: System Shock
30:05 DeepMind
32:20 AlphaGo
34:00 Cannibals and Cheaters
35:30 My AI-Generated Fever Dream
36:50 Chapter 5: The AI Doomsday Cult
39:41 Generative AI and ChatGPT
40:42 Sam Altman
41:40 Why LLMs and ChatGPT are vital
43:38 The End is Nigh! [or is it?]
47:14 Chapter 6: What Actually Scares Me
51:32 Final Thoughts
Thanks to Ground News for sponsoring: ground.news/ordinary
I’m fully convinced you’re a t-800. You can’t convince me otherwise.
Porn
Yeah, but can it strafe left in Crysis on high settings?
my adhd thanks you for making a boring subject with so many entertaining video clips
Can you talk about how AlphaGo was defeated by someone who never studied Go? That double sandwich strategy is the perfect proof that pattern recognition will never become intelligence regardless of how much data or training you put in.
What was the name of that jazzy slide guitar song around 6 minutes or so ?
What could possibly be a problem, if it isn’t already. Is in the case of AI with the purpose of guessing and suggesting content you want to consume. It would be reasonable for it to try to make people more predictable, as it would make its task easier.
…and extremist and people with radical black and white thinking are pretty predictable in what they want to see.
You know a video is about to be good when it contains so much niche old footage it’s only available in 720p
May the omnissiah have mercy on our fleshling souls
Nooooo Ordinary things got the broccoli hari cut nooooooooooooo
Climate scientists say China exceeded all developed nations combined in pollution. Source:BBC
I appreciate some of the humor and undercutting of the out there fears some people have about some real long terms issues with AGI, it is true that it’s easy to get swept up and let fear control your actions. At the same time, I do think it’s genuinely important to understand that the universe, reality, is not moral or ethical, atoms don’t think or care and even among humans morals are not inherent or widely agreed upon. Science tells us what things are not what they should be, and AI is similar, it can do logic and perform actions, but which actions it should choose to take is a question of morality humans have been trying to figure out since the beginning. It is true as well that there are other concerns that are a bigger problem right now though, certainly issues that will happen sooner.
I’ve been teaching my little brother to code and a point I find myself repeatedly reminding him is this: You have to tell a computer anything and everything you want it to do because it doesn’t just know anything the way we do, it only knows what you tell it. The same thing applies to AI, and we don’t know how to tell it to be moral, and while that concern is not as relevant right now, I don’t want people to underestimate that problem, it is really hard to solve, and we don’t have a solution right now. It’s not all fear-mongering, some of it, a lot of it, is real.
But again it’s scary stuff and I appreciate the humor we can’t find a solution if we are all running around scared or giving up, so thanks for this video :).
Hi Jreg