There is likely some tech that mirrors they way biology/muscles work and are powered that will be in our future, don't know the time horizon, it may only be a breakthrough or two away

OctopusBot 2030 begs to differ

I think google is full of smart engineers but terrible product managers, their products generally suck to use (Angular, Tensorflow etc) and they cancel a lot of them. The culture is a bit broken there.

I think people have this vision that Google is very capable and but the Emperor may actually have no clothes ...

No doubt their research is top tier (note that until now they didn't even merge their AI labs), but that is very different to releasing AI products. They could have released numerous products in this space, but they didn't.

Only now that they are forced to are they even trying and woe and behold they were unprepared. They got complacent and underdog sucker punched them to attention. Microsoft may have got lucky with their particular bet on AI but they are pushing very hard and it's great to see some competition.

I hope Google bring their A game but they have to beat Bing and GPT4 on price, quality and ecosystem, that may be quite challenging event with their resources

You have to admire the way they were willing to flip the table, if they didn't nerf bing they could have dethroned google relatively quickly

inequitbale distribution of suffering ;)

People think in increments and that's true to an extent but breakthroughs lead to big jumps e.g. chatgpt; robotics will have its chatgpt moment within a few years

They're just mad it's chinese, facebook, instagram, youtube, twitter, reddit all could be guilty of the same things as tiktok

We could also put our resources together and build our own ai, but seems while we can talk to eachother like this easily, we still cannot coordinate in any meaningful way

Robot Wrestling Manager

AI Therapist (AIs get sick of talking each other)

I think this is a bit premature because there is usually some form scarcity either by distance or time and a medium of exchange is necessary to trade these resources

e.g. gold is scarce localy but a sufficiently advanced space mining system will increase supply until we need to get out of the solar system, then you likely have to time constrained scarcity ie have to wait for operations in another star sysytem to be setup and send resources back

Even considering there are billions of people and perhaps a few very disirable places for them to live, how to allocate living space equitably since the qualities that make that space desirable cannot easily be scaled e.g. the historical / cultural value of living in Paris, the beauty of living in Hawaii

Why is it important to be positive and welcoming?

bitcoin?

what do you think of Boston Dynamics' Atlas robot?

do you mean highschool? I don't really see how highschool is anything but prison for young adults

This is an interesting idea, furthermore AI's that need to do things in the real world can use the same system, in advance of having robotics. More importantly, an AI using this can be acting via humans in many places at the same time

I think these models will soon be able to distinguish good from bad outputs and bootstrap their own improvement process

People said AI couldn't do art, write stories or make movies. These things are now either done or will be done soon. Robotics will go through the same progression, say within 5 years we have an android that can operate in complex environment, within 10 it replaces almost all manual labor

I think were now reaping the benefits of an increasingly networked society, news and advancements travel quickly and are iterated on by the smartest people, it's a collective intelligence that's only becoming more and more integrated.

What thought or real experiment would invalidate 3? You have to understand intelligence first to put system wide constraints on it like that, I don't think we can make those assertions.

You also have human evolution which came about in a low intelligence environment and rapidly gained intelligence, so I'm not sure why that would be different for machines.

Also on 1. you might say the brain is a type of computer, and it has nothing to do with Moore's law. Imagine we can replicate a similar system using synthetic neurons.

I think the problem is that if other jobs get automated out, there will be a lot of competition for the existing jobs which will drive wages into the ground.

So a future without UBI and sufficient new human jobs will be really bad for existing wage growth in existing roles as people retrain into whatever they need to.

When you stare at these generations for hours on end you can pickup subtle differences that give them away. If it's very good it's more like that show Severance, where they look at numbers on a screen and can tell some "feel" bad

I agree, I'd rather risk it all for a better outcome, the status quo sucks

They seem behind the ball, openai has so much interaction data now