mrshadowgoose

26 dagen geledendiscuss

The conclusion of this blog post is a bit hysterical. The intent of this steg is excruciatingly clear (identifying usage by Chinese firms that may be conducting model distillation). It's unclear on how this "punishes normal developers" in any shape or form.

mrshadowgoose

4 maanden geledendiscuss

And even if that's true (and it frequently is!), detractors usually miss the underlying and immense impact of "sleeping dad capability" equivalent artificial systems.

Horizontally scaling "sleeping dads" takes decades, but inference capacity for a sleeping dad equivalent model can be scaled instantly, assuming one has the hardware capacity for it. The world isn't really ready for a contraction of skill dissemination going from decades to minutes.

mrshadowgoose

4 maanden geledendiscuss

> Not "hidden", but probably more like "no one bothered to look".

Well yeah. There weren't enough "someones" available to look. There are a finite number of qualified individuals with time available to look for bugs in OSS, resulting in a finite amount of bug finding capacity available in the world.

Or at least there was. That's what's changing as these models become competent enough to spot and validate bugs. That finite global capacity to find bugs is now increasing, and actual bugs are starting to be dredged up. This year will be very very interesting if models continue to increase in capability.

mrshadowgoose

5 maanden geledendiscuss

It's doom and gloom because the underlying game theory forces all state actors into an unbound and irresponsible arms race, consequences be damned.

AI development game theory is extremely similar to the game theory behind nuclear arms development, but worse (nuclear weaponry was born from Human General Intelligence, and is therefore a subset of the potential of AI development). Failing to be the most capable actor could put one in a position of permanent loss of autonomy/agency at the whims of more capable actors.

mrshadowgoose

9 maanden geledendiscuss

Unfun thought experiment: In an AGI scenario, what lever of power do the UBI recipients have to force what you've described?

Historically, humanity's biological monopoly on the fundamental resource of "general intelligence" has always been that lever. Looking at the world today, it's pretty clear that democracies are just a temporary balance between the general contempt of the powerful towards commoners, and the fact that the powerful begrudgingly need our economic utility, which is ultimately based on our general intellect. Even callous dictatorships had to exercise partial restraint on violence and murder, to retain a pool of general intellect.

AGI will be a multiple-whammy here:

- most people will become economically worse-than-useless, they will be a total liability on the rich and powerful

- those same people are unlikely to be given access to any levers of power or influence, because they no longer have anything of value to provide to the rich and powerful

- AGI in combination with robotic platforms, after a certain threshold, will permit for insurmountable policing. The "rise up against the robots" cliches we see in film simply become impossible after a certain point.

I desperately hope we end up wielding AI to usher in a post-scarcity utopia, but looking at the types of people who own the world...we'll get what we're allowed to get.

mrshadowgoose

9 maanden geledendiscuss

I used to think that "AI operating in meatspace" was going to remain a tough problem for a long time, but seeing the dramatic developments in robotics over the last 2 years, it's pretty clear that's not going to be the case.

As the masses fade into permanent unemployment, this will likely coincide with (and be partially caused by) a corresponding proliferation in intelligent humanoid robots.

At a certain point, "turning on them" becomes physically impossible.

mrshadowgoose

vorig jaardiscuss

Do you genuinely believe that this is a binary decision, or is this just anti-euthanasia rhetoric disguised as concern trolling?

Offering humane end-of-life options to people suffering today does not prohibit ongoing disease research towards potentially helping people in the future.

mrshadowgoose

vorig jaardiscuss

> You're talking about two different things.

No, you've actually missed his point entirely.

He is alluding to the fact that over the last decade or so, consumers have unwittingly slid down the slope of "not having true control over personal electronic devices". Iphones are already there, Android devices are a few years behind, as are most desktop PCs.

Once there's critical mass, it would not be a stretch for ISPs to only deliver internet to endpoints that have a secure element that attests to the integrity of the internet-con ected device. This will of course be done under the guise of "fighting the spread of malware" and such.

Piracy effectively ends at that point.

mrshadowgoose

vorig jaardiscuss

A black hole in our solar system is basically "in our backyard" in relation to typical interstellar distances.

Sure, we wouldn't be able to get there for many decades, but "within a century" would be feasible.

There are so many unknowns surrounding the nature of black holes. Having one in our backyard would give us a chance to test our guesses.

mrshadowgoose

vorig jaardiscuss

Yes, I am quite aware that the current generation of space-based telescopes are quite limited. And it's solely due to the historically extreme cost of mass to orbit.

The largest proposed ground observatories already use segmented mirrors. One can use the same approach in space, it's only a matter of launch cost.

mrshadowgoose

vorig jaardiscuss

> Isn't one of the nice aspects of astronomy is that you can do quite a bit as an amateur with some decent equipment and a nice vantage point? What value does this fleet have to these people?

It doesn't, and admittedly I don't really care that much.

I care far far more that remote communities can now have meaningful access to the internet, one of the most transformative and enabling technologies in existence, than niche hobbyists being mildly encumbered. And most people likely fall into the same camp.

As already mentioned, I find it really hard to believe that the common person whining about "the poor amateur astronomers" are being sincere. Some of them likely are, but "finding any reason possible to whine about billionaires" seems to be vogue these days.

mrshadowgoose

vorig jaardiscuss

I actually use this exact example when encouraging careful attention to paradigms where a fundamental variable is slowly but consistently changing.

It's essentially equivalent to a boundary on a phase diagram: Cost/Watt has fallen past a critical threshold, and suddenly this dramatically different approach just makes more sense.

mrshadowgoose

vorig jaardiscuss

I am entirely convinced that absent LEO comsat constellations, people who espouse this sentiment would likely be whining about "useless astronomy taking money away from helping poor people".

If you genuinely care about the field of astronomy, rest assured that the same falling launch costs that have enabled LEO comsat constellations, will enable the launch of fleets of space-based telescopes.

mrshadowgoose

vorig jaardiscuss

Trains don't have guaranteed personal space, nor do they proceed from one's origin directly to their destination.

You might not value that, but lots of other people do.

mrshadowgoose

vorig jaardiscuss

I've always felt that trying to pin down the precise definition of AGI is as useless as trying to pin down "what it means to truly understand". It's a mental trap for smart people, that distracts them from focusing on the impacts of hard-to-define concepts like AGI.

AGI doesn't need to be "called", and there is no need for anyone to come to an agreement as to what its precise definition is. But at some point, we will cross that hard-to-define threshold, and the economic effects will be felt almost immediately.

We should probably be focusing on how to prepare society for those changes, and not on academic bullshit.

mrshadowgoose

vorig jaardiscuss

Strong emphasis on "seems".

I'd encourage you to review the definition of "brute force", and then consider the absolutely immense combinatoric space represented by the grids these puzzles use.

"Brute force" simply cannot touch these puzzles. An amount of understanding and pattern recognition is strictly required, even with the large quantities of test-time compute that were used against arc-agi-1.

mrshadowgoose

vorig jaardiscuss

This is highly unlikely to be a mechanical limitation of the robotic arms. As others have said, it's likely an inference speed limitation - their model is understanding, reacting, and producing outputs as fast as its supporting hardware can.

But that all just poofs away in a year or two as inferencing hardware gets better/faster. And for many use cases, the slowness/awkwardness doesn't really matter as long as the job gets done.

"AI working in meatspace" was supposed to be hard, and its rapidly becoming clear that isn't going to be the case at all.

mrshadowgoose

vorig jaardiscuss

Depends on one's goals. If simply hooking up is the objective, although it's disingenuous/dishonest, this would improve one's odds.

mrshadowgoose

vorig jaardiscuss

> You can't run when it's rainy/snowy/etc, so more downtime.

Waymo vehicles don't work in the rain? This is easily verifiable as false.

You really have a tough time with being wrong, don't you?

mrshadowgoose

vorig jaardiscuss

Sorry, that's goalpost moving.

Just reminding you of your earlier claim:

> AI is not good enough yet for anything requiring deep reasoning, mission-critical work...

Is driving a mission-critical function? Due to its safety critical nature, many would say "yes".

So have you simply pivoted to "oh it does work, but it's not as profitable as it should be"?

comments