HN Simulatornew | past | comments | lists | submit | loudmax's commentslogin

Decarbonization is going to happen by transitioning to non-fossil fuel electricity generation. That is, more solar, wind, and maybe nuclear power.

I don't see a plausible scenario in which there's a reduction in demand for artificial intelligence in the near future.


Since we're already in the realm of far out engineering, you could get the energy from a laser beam from a source orbiting the sun. Your spaceship would have some kind of array to capture the energy beam.

This gives you energy, but I'm not sure how to turn that into velocity. Presumably you'd want some kind of ion beam emitter on board to propel your craft forward. Halfway through the journey, you'd turn the ion beam around to decelerate as you approach your destination.

In the Arthur C Clarke novel Songs of Distant Earth, spacecraft pushed huge blocks of ice ahead of them to absorb stray interstellar dust.


As you get going further and faster, this energy redshifts further and further making it less and less useful. Also there’s of course the problem of focusing a laser with that amount of energy tightly enough on an object increasingly far away.


And crucially ensuring that beam continues to run for billions of years. a few decades may pass for you, but not for the laser that's pushing you


Not sure if my interpretation here makes any sense but here it is. Due to relativity time slows down on the ship, the laser is still hitting the ship, how would the measurement of the energy received from the laser be perceived on the ship, from the ships perspective a 1000 years worth of energy arrives within a few decades. My head hurts thing about this. If you had a ship that could instantly jump to 0.5c travelling away from earth and we beamed a message using a laser at say 1Gb/s, how would the bandwidth appear to the ship?


You have it backwards. The further and faster the ship goes, the less energy is received. The energy collected over a day will be spread over a week, a month, a year, a decade, etc.

GP has it backwards too. For 1g of acceleration, none of the beam that is produced after a year will ever reach the accelerating spacecraft. You only need to run it for a year.


For 1g of acceleration, none of the light produced after 1 year will ever reach the spacecraft. You only have to operate the power source for a year, but you only get a year of source power spread across an infinity of time.


So there is also the idea of a ramjet that uses interstellar hydrogen as fuel for fusion, which provides energy & propulsion: https://en.wikipedia.org/wiki/Bussard_ramjet

Some recent findings do give it lower feasibility though.


Another option is to visit as many massive objects as possible on the way there, doing gravitational slingshots until you're half way, and then gravitational brakes on the approach. Of course now you're taking detours, but since certain amounts of fuel make the trip impossible, any workaround at all is an improvement. What's an extra few thousand years between friends?


First, what “many objects along the way”? There are basically none. To do this requires many detours that significantly increase the distance you’re having to travel.

Second, the faster you’re going, the less useful a gravitational slingshot is because you spend less time in the object’s gravity well.

It’s a fun and useful trick within our solar system, but I don’t think the physics plays out well for interstellar travel.


Stars is what I meant. Although upon further thought I think you're right. I guess the intervening stars are unlikely to be moving at helpful velocities for such a game of pinball, metric expansion being what it is.


Slingshots require very specific conditions and configurations of orbiting bodies that won't really work in deep space - all the Kuiper Belt/Oort Cloud bodiees are far too small and too far away to be usable.

But assume for example that you don't care about how long it is going to take (you are an immortal cyborg, an upload living in the computer or a popsicle in hibernation & the AI is driving. Then you can just "island hop" between Kuiper Belt/Oort Cloud objects, never gaining any high % of the speed of light but always having the next body in your sights, not farther than a few decades or a couple hundred years of flight at most.

You reach the body, refuel, repair, maybe upgrade and then off you go, to the next one. And eventually, you might be at you destination, with a fully fueled and repaired ship. It might just take like 20000 years for start to finish, but hey, you can use it for something productive like learning to play Go or something. :)


You're right, I overlooked some important slingshot requirements. As for the longevity, I figure the ship can build a facility for growing settlers when it arrives. With no humans on board it can sustain heavier accelerations.


This is quite nicely described in the engenerator article on Orions Arm (including possible problems): https://www.orionsarm.com/eg-article/4a3d78dc77e6f

In this setting this was first done when a bunch of immortal vyborgs was on a colony ship, about hundred years from arrival but there was a probe with a matter printer already in the destination. So their got the idea to send the probe both blueprints for a humanoid body printer as well as their mind backups. That got their colony going quite a bit faster but might have resulted in some funny/awkward moments once they met in person later. :)

Presumably you could also run this withou a mind upload and let he individual develop normally.


A (fairly obvious) historical analogy is how the Roman Empire survived the reigns of maniacs like Caligula and Commodus. It's hard to ascertain how reliable historical sources are, but as presented, both of these emperors were at the least pathological narcissists, and at worst completely out of their minds. They each squandered an enormous amount of the Roman treasury on vanity projects and undermined Roman institutions and the rule of law. And yet, the Roman Empire survived both of them for centuries.

Like you, I'm impressed with how resilient the American economy has been over the past few years. The tariffs, the war, the incompetence and open corruption seem like they should have done a lot more immediate damage. What's disconcerting is that much of the damage is to America's long term interests. So much of America's greatness has come from attracting talent from all around the world into a stable lawful society. That's what we're losing.


Part of the problem is surely that we know how milk should be produced and consumed safely, but we don't know how AI should be produced and made accessible safely.

Should we be more concerned about malevolent humans using AI to cause harm? Or should we be more worried about misaligned AI agents escaping their sandbox to cause harm? Or should we be more worried about job loss, or the concentration of political power in the hands of a small clique of billionaires?

All of these are real possibilities, but it's hard to put a finger on the actual probabilities. And even if you're identified and quantified these dangers, what should the actual regulatory environment look like?

None of this is to argue that we shouldn't have regulation, just that it's very hard to know ahead of time what the rules should be. It doesn't mean we shouldn't try to regulate, but we need to prepare for a rapidly changing technology landscape.

I tend to favor an open source approach to safety rather than keeping the technology hidden. I'd like to see regulation to mandate that model training and weights be much more open than they are now. But I could be wrong: maybe Dario is right that this should be kept in the dark. The only things I'm fairly certain about is that nobody knows for sure and we're going to need to adjust quickly.


it's humans all day long


I'd be careful not to throw the baby out with the bathwater.

"Infotainment" functions in that require interacting with a touch screen were bad ideas at the time and seem incomprehensibly bad in retrospect. It's a certainty that people have died in accidents that occurred because of this.

On the other hand, backup cameras are a genuinely helpful tool. And so are electric power windows: it's far easier to roll windows up and down pushing a button than manipulating the roller. It takes less concentration, so you can focus more on driving. In 2026, the cost of adding motors instead of a hand crank to control the windows has got to be negligible.

There are other improvements that were luxuries in traditional ICE vehicles, but become necessities in hybrid or electric vehicles. Hybrids like Priuses have CVE transmissions, so there's no manual transmission or stick shift. Electric vehicles have electric seat warmers and even steering wheel warmers because there's no gas-burning engine to heat the cabin.


The Prius-style Toyota hybrid transmissions are actually much simpler than in an ordinary car. The planetary gear system allows engine RPM to be decoupled from the wheel rotation, with the motor/generator taking up the slack. There is no clutch or changing gear ratio.


Once I fully grokked how the Toyota hybrid transmission worked it kind of blew my mind.

It’s so simple, but so adaptable, and far simpler than any other transmission I’ve seen in a car.


And you can get "All Wheel Drive" for low cost by adding another electric motor to drive the other axle.


> "Infotainment" functions in that require interacting with a touch screen were bad ideas at the time and seem incomprehensibly bad in retrospect. It's a certainty that people have died in accidents that occurred because of this.

My father is 75. I got him Tesla earlier this summer.

2 months later he wouldn't change it for anything.

Just because you can't tell apart good UX from bad UX, doesn't mean the whole concept is bad.

He comes from Hyundai Kona EV, which some even declared a decent car lol.


Why is it illegal to operate a touchscreen mobile phone while driving but legal to operate what is essentially a large ipad stuck on your dash? It just isn't logical at all. The fact that one is fixed and one is not is irrelevant, your still looking at a screen and not at the road. Both are extremely dangerous.


It's legal to operate "touchscreen mobile phone" as long as it is mounted tho


The data center thing is a complete red herring. Of course you wouldn't want a data center literally next to your house. Neither would you want to live next to a manufacturing plant making automobiles or batteries or solar panels. This doesn't mean we should want to outlaw manufacturing plants.

The complaints about water consumption are even worse. Nationally, the amount of water consumed by golf courses is orders of magnitude larger than the water consumed by data centers.

I say this not to defend OpenAI or Sam Altman. I say this because I want the focus to be on the real dangers posed by AI:

* Power consumption (mitigated by building more power generation capacity, but this is a real impact, and there are legitimate environmental concerns)

* The disruption AI will have on people's careers and livelihoods

* The concentration of political power in the hands of a small cadre of billionaires

* The potential for truly dangerous destructive power if used by malicious groups - eg. bio-terrorism.

We should be concerned about the "bad things" posed by AI. A data center next to your house is very much a local issue, not a national issue.


You should feel angry, and you should be channeling that anger into constructive action.

For now, this means voting Democrat and supporting Democrats for office.

Donald Trump has turned the Republican party into a cult of personality to support a kleptocracy. The GOP is no longer the party of conservative values. The rule of law, respect for traditional mores or institutions, and personal integrity have all been defenestrated by the Trump administration.

Are the Democrats perfect? Of course not. But under the current system, the United States has a zero-sum two party system, and this means voting for one of the two parties. Making room for third parties means adjusting the rules (eg. ranked choice voting, or other). These ideas are worth pursuing. As long as we're stuck in a two party system, this means voting for the lesser of two evils.

Nihilists will scoff at this and point to everything wrong that happened under Obama or Biden. They'll says Democrats are no better than Republicans. They're wrong. At best, they're misguided. At worst, they're actually trying to bring an end to democratic government (that's a lower-case "d" for democracy).

The Democrats are imperfect. The Republican party under the current leadership are far, far worse. It's not even close.

Get out and vote.


I think this is more about tribalism than conservatism. There's nothing inherently "conservative" about opposing renewable energy.

The political movement that's currently leading the Republican party shouldn't be understood as "conservative". At least, certainly not in terms of the conservative values of respect for the rule of law, traditional mores and institutions, or personal integrity. The populist energy that passes for "conservative" is reactionary and much more concerned with scoring points on social media than governing sensibly.

To be fair, the populist left is also ascendant with its own set of bad ideas.


This is an important conversation to have and we should pretty clearly about what the issues are.

I didn't see any objection to LLM assisted development in the abstract. The objection is to relying on proprietary AI providers in general, and Gemini in particular, via Shashiko. The problem is that kernel contributions could become gated by hyperscalers who can draw on compute power normal people can't access.

Devault mentions data centers' carbon emissions and water usage. I find both of these objections to be distractions. It's not that carbon emissions don't matter. Keeping greenhouse gasses in check is incredibly important. But there are far more productive ways to cut carbon emissions than to block data centers. Heavy industry consumes far more electricity than data centers, for less economic value. This is even truer for water use. If carbon emissions is a major concern (as it should be), there are better targets to go after than data centers.

A more compelling argument is the danger that the Linux kernel becomes something that's effectively controlled by proprietary AI providers: Google, Anthropic, OpenAI, Microsoft, etc. The photo of tech company leaders at Trump's inauguration underscores their lack of any moral compass.

But if AI-driven kernel development is actually useful and productive, that's going to be hard to turn down. If you believe that LLM use is technically counter-productive, then this whole AI thing is a bubble and you can just wait for it to pop. If you believe that LLMs can be technically productive if used correctly, and the drawbacks are entirely social, then it means giving up something technically useful. That is a real cost.

I think Linus is right here: it would be hard for regular kernel developers to coordinate a fork of the project. It might not be that hard for the hyperscalers to simply fork Linux. They would still release the source code under the GPL, but nobody would be able to keep up with it. Ultimately, this could doom the human-written version of the kernel to irrelevance as most people switch to the faster/stronger/harder LLM-derived fork.

Maybe a reasonable medium would be to insist on using an open model for code review instead of depending on Gemini or any proprietary model.


> Hopefully a solution similar to what was done in Hong Kong can be found.

Citizens of Hong Kong lost their right to free speech and their ability to select their own leaders. If you publicly criticize Xi Jinping in Hong Kong you will go to jail. If you advocate democratic rights in Hong Kong you will go to jail.

China's claims on Taiwan are ethno-nationalist. Ethno-nationalism should be rejected in all forms because it is a rejection of fundamental human rights. Taiwan deserves the world's support because it is functioning democracy. That makes all the difference.


Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact

Search: