@Magcheck
This is an important conversation we need to have but NBC - lose the the freakin background music. By including it you've managed to dumb down the importance of this very serious issue.
@thomassimeroth701
2:14 AI so speedy it added an extra E to SPEEED
@alcapone1486
If you've been looking for an educational video with important conversation and with awful background music that makes the video irritating, this is it.
@FISHESROCK101
Omg the music In the background makes it Impossible to focus on what he's saying
@THOMMGB
I've got to wonder, where are we going to be in 5 years?
@tinfoilslacks3750
There are huge problematic ramifications of generative algorithms (I'm choosing to use the term generative algorithm because not all AI generate content, and the problems I'm talking about are unique to these models which output content, not a problem with the mere existence of a neural network). 1. The direct appropriation of creators to replace them. These algorithms create new content which is a learned approximation of existing content which is given to them to learn from, called the model. These can be so specific that an algorithm can, for instance, comb through its model and create new content that directly mimics and replaces an existing specific artist or creator, not just creators in general or conceptually. You can ask these things to say, "create an image of X in the style of Y artist" and it will. Essentially replacing a creator by virtue of actively appropriating their creations. 2. Misinformation and Imitation. These algorithms can create content depicting real people, both living and dead, in terms of video and audio. You could pass off created content as legitimate content of a real person for the sake of misinformation or slander or even political manipulation. You could also, for instance, use these generative algorithms to create "new" content of dead artists and creators and directly profit from it. The "ghost" of a past creator can now exist in perpetuity at the command of a corporate entity for profit. 3. Ambiguity and saturation. As more outlets adopt bots and their outputs, and as the output of these bots improves, you will be less and less capable of telling bot content from real content. You could be watching a YouTube video and be totally unable to tell which comments are real or generated. Chatbots and spambots and scams can be indistinguishable from real people and used for nefarious purposes. Every interaction with any content will have to be done with the understanding literally anything you see could have been bot generation. The further these develop, the greater saturation of bot content online there will be. You'll have to navigate online understanding everything is potentially fake, including, potentially, something that right now seems like it can't be faked like a real time video call. 4. Huge wealth inequality will become even more disproportionate. With just a few generative models able to replace literally entire industries, the corporate entities who control the models and the corporate entities who own the means of broadcasting and profiting from these creations will accrue untold wealth that currently circulates through the millions of people who work in these industries. 5. direct obsolescence of creative pursuits and cultural enrichment. In a world where writing, art, music, and video are automated, any manual work in these fields will be entirely unprofitable which will directly cause these professions to diminish considerably. When simple automation or dangerous heavy labour happened throughout the twentieth century, it directly benefitted the majority of people by reducing dangerous work and increasing productivity. This form of automation will automate the type of enriching and culturally/socially affirming work increasing productivity via machine automation in the twentieth century is supposed to propagate. This automation makes human endeavours that enrich our life obsolete, and ironically leaves us only with the manual labour we've been trying to escape to pursue these more engaging activities. The technology itself is fascinating and impressive. However, it's the sort of technology that regardless of how impressive it is has massive negative consequences for the vast majority of people and only very fringe minimal benefits.
@Gwynarra2
It’s interesting to note that this segment has much less views and comments than the school shooting segment. Both segments were released within an hour of each other but clearly one subject is getting more views. I wonder what AI would do with that insight into human nature.
@Recuper8
Omg, the background music must go!
@juliecornfield3590
is there a version of this video without the needless ~futuristic music~ in the background
@MicahBratt
We know that the human brain uses previous data to understand new data, finds patterns and uses a system of rewards for developing new behaviors. These LLMs seem to be pretty good at modeling the basics. So what's the missing, the quantum factor?
@2oon24
Great video but background music should be much louder
@Slayr.
There should've been better music.
@ka9dgx
Holy cow, I forgot how good broadcast level production is. How many man-hours went into this, total?
@deusbeowulf6039
Almost like saying “more interstellar space flights are happening every month. How do they work?”. Some things cannot be explained in under ten minutes lol.
@georgemaximus694
It’s game over for humanity. We have created what will replace us completely… revolutionary indeed… like humans replaced the dinosaurs.
@pikasnipe1
It's over before It's begun.
@HardKore5250
I made it write Freddy Kruger pop tart commercial funny😂!
@MHowieDoC
Interesting, but as a child of the 80's I have thought about AI 🤨 since Skynet in Terminator 2.
@teragram8006
this is on par with the nuclear bomb. development and training shouldn't be happening in closed off labs. it should be happening openly for all humanity to observe. the models are being trained off of actual human interactions and content we've all created. everyone has a stake in this
@AIwithOliver