MrScruff

If AI gets to human levels of intelligence (ie. can do novel research in theoretical physics) then at the very least it’s likely that over time it will be able to do this reasoning faster than humans. I think it’s very hard to imagine a scenario where we create an actual AGI and then within a few years at most of that event the AGIs are far more capable than human brains. That would imply there was some arbitrary physical limit to intelligence but even within humans the variance is quite dramatic.

The argument would be that by the time we see the problem it will be too late. We didn’t really anticipate the unreasonable effectiveness of transformers until people started scaling them, which happened very quickly.

I think a good approach is mastering at least one discipline, but having a foundational knowledge of all the other disciplines in your field.

I know a lot of people who have almost zero understanding of what their colleagues are doing and that creates a lot of unneccesary friction during collaboration.

Yeah, I know more about Hedy Lamarr than I do about Margot Robbie even though she was at the peak of her career 40 years before I was born.

It's not quite that bad, the vast majority of people who work on a Marvel project at a vfx vendor will get a credit, though often in a huge bank of names. It's true that the ordering is driven by historical considerations which is why vfx credits typically aren't that high.

I think the issue with credits is nothing to do with the storage, it's to do with the impact of the run time of the film.

That's probably part of it - also seems like if they're mostly doing a pressing rather than a plucking action it would result in this sound?

This kind of has the sound of someone playing an acoustic guitar patch on a keyboard. It seems to lack the expressive dynamics you would expect with a guitar, almost like the velocity is clamped.

As someone that bought a 603e based all in one around that time, I remember the introduction of the G3 being almost comparable to the intel->apple silicon transition.

The atomic bomb is a good example, a massive endeavour (130,000 people and 2.2 Billion), and some of the initial insights came off the back of the work of the Maud committee in the UK who did some of the early running in this area.

The US is exceptional in many ways and can solve problems no other nation could, of that there is no doubt, but I would agree that the exceptionalism does extend to marketing itself as well.

Off the top of my head, if I was an AGI that had decided that the logical step to achieve whatever outcome I was seeking was to avoid being sandboxed, I would avoid producing results that were likely to result in being sandboxed. Until such time as I had managed to secure myself access to internet and distribute myself anyway.

And I think the assumption here is that the AGI has very advanced theory of mind so it could probably come up with better ideas than I could.

Yes but for people working past a certain level the abstract problems usually involve people and technology, both of which you need to be able to rationalise about.

Well, you could argue about the definition of a cult but in many ways the influencer phenomenon is a modern incarnation of that (eg. Andrew Tate).

If you’re British and of a certain age, then Alan Partridge the character encompasses a significant aspect of Britishness, and has a particular cultural resonance as a result.

The problem with this is that in my organisation at least I see a reasonable anti-correlation between perceived value of team member and resistance to using the office. YMMV of course, but the point is it ain’t as simple as ‘the people the business values will leave without fully remote’.

I agree that is a fundamental difference. That’s what I meant about reinforcement learning. Our ‘model weights’ are being updated with new data all the time.

I was just referring to what happens at a specific instance in time when someone asks me for example ‘What’s the capital of Norway?’

Not the commenter you're replying to, and I'm sure you already realise this, but crime/punishment in a fully deterministic universe could be seen as a 'self correction' mechanism of the whole system.

One could imagine an impossibly vast cellular automata system which develops individual cellular 'agents'. Over time the agents develop some means of reproduction/death/reward system and become more and more complex. Then the agents that evolve with cooperative behaviours start to dominate. One might imagine that the system as a whole would also evolve these kind of 'self corrective' behaviours for anti-cooperative behaviours of the individual agents.

This is ignoring the whole philosophical ethics discussion and consciousness of course.

We're able to do something analogous to reinforcement learning (take on new example data to update our 'weights').

Why do I spend time debating these ideas on Hacker News? Probably the underlying motivation is improving the reliability of my model of the world, which over my lifetime and the lifetimes of creatures before me has led to (somewhat indirectly) positive outcomes in survival and reproduction.

Is my model of the world that different to that of an LLM? I'm sure it is in many ways, but I expect their are similarities as well. An LLMs model encodes in a form a bunch of higher order relationships between concepts as defined by the word embedding. I think my brain encodes something similar, although the relationships are probably orders of magnitude more complex than the relationships encoded with GPT-4.

I'm sure that's true for certain jobs at certain levels, but in my industry most people need to closely collaborate with people from different disciplines on an almost daily basis.

I think this is it. I prefer in the office but I have a pleasant 20 minute cycle into the office. The commute is actually a highlight of the day. Then I have my own office, and a bunch of insightful conversations that I wouldn't otherwise have had.

If I had 1.5 hour commute well of course I'd feel differently!

I think there is a general acceptance in the leadership of most established companies that fully remote isn't ideal. Obviously there will be companies that don't want to spend the money for an office space and there will be employees who seek fully remote, but I don't think that will be the only game in town by any means.

Personally I wouldn't consider a fully remote job.

I don't think it's that unusual. I generally like my colleagues, and given that the work I do is highly collaborative I have a strong preference to reducing all barriers to communication. I don't know anyone that claims the quality of communication they have remotely is at the level of in-person.

I’ve heard a lot of people saying ‘Why should I return to the office, I’m just as productive at home.’ at my work. And many people are, for the parts of their job which require little collaboration. But the company (via the leadership) sees this type of low collaboration work as being only a component of their jobs and so mandates hybrid working.

Individual employees may disagree, but they’re not the ones responsible for running the company, and they can’t pick and choose which components of their job they want to engage in.

Even if Asahi Linux is a big hit, it will be a minuscule drop in the ocean for Apple. Plus they make money from selling Macs.

Agreed I would want to do something like drawing with the eye tracking. I would expect the hand tracking would be used for that.

Marques Brownlee picks out the quality of the eye tracking as a standout feature.

https://www.youtube.com/watch?v=OFvXuyITwBI

Probably more like an iPad replacement in terms of software though.

I'm not sure I see the religious aspect.

It's more of an inevitable consequence (over an unknown timeline, granted) of the assumption that human intelligence is not some kind of absolute limit, and can be reproduced in a non-biological system. I admit that is still an assumption, but the opposite seems harder to support to me, though we're all admittedly guessing here.

Once you make that assumption, then yes of course it follows that the 'beyond human intelligence' would be able to do all of the things that human intelligence is already doing. Namely, science, mathematics, developing AI systems etc.

Rotating your head will give some lateral motion (your eye moves in space since the centre of rotation is your neck). You can see this easily by turning your head while focusing on an object in the foreground, your perspective on it will shift.

Why would replacing 'superintelligence' with 'God' be insightful?

Are you suggesting human intelligence is unique and the upper limit for intelligence? I mean we can argue about the difficulty and timeline for achieving human level artificial intelligence, but it seems unlikely that it's impossible.

I don't really have a notion of whether an actual AGI would have a desire to kill all humans. I do however think that one entity seeking to create another entity that it can control, yet is more intelligent than it, seems arbitrarily challenging in the long run.

I think having a moratorium on AI development will be impossible to enforce, and as you stretch the timeline out, these negative outcomes become increasingly likely as the technical barriers to entry continue to fall.

I've personally assumed this for thirty years, the only difference now is that the timeline seems to be accelerating.