All of knb's Comments + Replies

I think the situation is broadly analogous to advanced nuclear reactors. Both have major strategic military, foreign policy and economic applications.

  1. With high temperature reactors you can efficiently synthesize liquid fuels. Energy security is a major strategic goal of US foreign policy.
  2. The US developed nuclear powered airplanes to a high level. Imagine strategic bombers that could fly continuously for weeks striking targets anywhere on earth without refueling or needing nearby airstrips.
  3. Supersonic nuclear ramjets like Project Pluto were feasible as w
... (read more)

I think we’re currently in an era of unusually large amounts of free speech that elites are starting to get spooked by and defend against.  Most people have high, perhaps even growing, tolerance for controversy and offense, but some find it unacceptable, and these people are disproportionately influential.


People always had a lot of free speech, it was just in unmediated human interaction. There was a lively civil society and people had closer relationships with neighbors and extended family and would naturally discuss things. This decentral... (read more)

I'm not sure if the agenty tasks you have in mind are considered part of manufacturing per se or business management. My impression from above is that production work and factory construction is being automated but design/engineering and business management are not. I'm not sure, but it does seem likely that humans could be out of the loop without AGI. (Though of course AGI could happen before narrow AI actually achieves this level in practice). 

I don't know, but I would guess that 90%+ of the industrial/manufacturing jobs done by humans even just 50 years ago are now done by machines. But this 90% automation didn't lead to a near-doubling of GWP growth rates.

90% automation only gives a ~10x increase in per-worker productivity in manufacturing. Since manufacturing is only a fraction of GWP, a 10x productivity increase only makes GWP (per capita) a few times larger. Take humans out of the process completely and the bottleneck is gone. The feedback loop is only constrained the availability of resources. 

2Daniel Kokotajlo3y
Good point. I guess I just find it implausible that humans will be COMPLETELY out of the loop prior to AGI. Some parts of the loops involve agenty tasks in which you draw on general world-knowledge to make novel plans and strategies and then execute them learning and adapting constantly.

Looks like it is still the global reserve currency.

Unless I'm missing something, looks like #1 is actually correct... 2019 GDP PPP per capita for Singapore was 103,181 according to IMF, which adjusts to 84775.10 according to the first inflation calculator on Google.

What did you mean by "fanon dwarves"? Is that just a fan interpretation or do you think Tolkien intended it? In Tolkien's idealized world, all economic motivations are marginal and deprecated. The dwarves are motivated partially by a desire for gold, but mostly by loyalty to their king and a desire to see their ancestral homeland restored to them. To the extent the treasure itself motivates Thorin & co., it causes disaster (for example his unwillingness to share the loot almost causes a battle against local men & elves.)

So it causes disaster. Disaster on this scale was already likely, even warned against. Somehow, the elves inviting themselves along - out of pure greed, no magic involved - is portrayed as less of a fault.
Thanks! This was what I had in mind!

I think it's mainly about combining two click-friendly buzzwords in a novel way.

This is a good example of the type of comment I would like to be able to downvote. Utterly braindead political clickbait.

The fact that journalists at a mainstream publication use the metaphor of machine learning to explain the actions of the president is noteworthy. Five-years ago you would be hard pressed for a journalist who thinks that his audience would understand machine learning enough to get the metaphor.

One thing to watch for would be top-level AI talent getting snapped up by governments rather than companies interested in making better spam detectors/photo-sharing apps.

This seems relevant These statistics do not support your claim that energy consumption per capita has been stagnant. Did I miss something?

Yep, your link is for world energy use per capita, my claim is that it was stagnant for the first world. E.g. in the US it peaked in 1978 and has since declined by about a fifth. Developed world is more relevant because that's where cutting edge research and technological advancement happens. Edit: here's a graph from the source you provided showing the energy consumption history of the main developed countries, all ... (read more)

Like a lot of reddit posts, it seems like it was written by a slightly-precocious teenager. I'm not much of a singularity believer but the case is very weak.

"Declining Energy Returns" is based on the false idea that civilization requires exponential increases in energy input, which has been wrong for decades. Per capita energy consumption has been stagnant in the first world for decades, and most of these countries have stagnant or declining populations. Focusing on EROI and "quality" of oil produced is a mistake. We don't lack for sour... (read more)

This seems relevant These statistics do not support your claim that energy consumption per capita has been stagnant. Did I miss something? Perhaps you're referring strictly to stagnation in per capita use of fossil fuels? Do you have different sources of support? After all, this is merely one data point. I'm not particularly sure where I stand with regards to the OP, part of the reason I brought it up was because this post sorely needed evidence to be brought up to the table, none of which I see. I suppose this lack of support gives a reader the impression of naiveté. but I was hoping members here would clarify with their own, founded claims. Thank you for the debunks, I'm sure there's plenty of literature to link to as such, which is exactly what I'm after. The engineering behind electric cars, and perhaps its history, will be a topic I'll be investigating myself in a bit. If you have any preferred sources for teaching purposes, I'd love a link.

Speaking of which, can anyone recommend any short, intelligent, rational writings on feminism for instance? My average exposure to anti-feminist thought is fairly intelligent, while my average exposure to pro-feminist thought is "How can anyone disagree with me?[...]"

There are some intelligent and interesting heterodox feminists who spend a lot of their time criticizing mainstream or radical feminist positions. I could recommend them to you, and you would probably like some of what they have to say, but then you wouldn't really be challenging... (read more)

Feminist thought: "Men have power and that's sexist. Smash the patriarchy!" Anti-Feminist thought: "Feminism has too many bad eggs". Feminist thought: "You can't say that, you racist sexist bigoted bigot!"

Isn't a VAT already basically a Robot Tax?

Seems less so than a tax on corporate profits is. Am I missing something?
That would explain all those sci-fi robots who only walk around destroying stuff and never build anything. They were programmed with an incentive to keep the VAT low, they took it too literally, and things got out of control.

The same game theory would seem to apply equally well in both cases. In what way does it work better with climate change?

I think it's a clear enough prediction, but putting some actual numbers on it would be useful. Personally, I would put the odds of a Trump landslide well under 50% even contingent on "supercharged" economic growth. Maybe 25%. Politics is too identity-oriented now to see anything like the Reagan landslides in the near future.

Kudos for making a clear prediction.

I voted for Trump but I don't think there is any realistic possibility of a Trump landslide, even if the economy grows very well for the next 4 years. The country is just too bitterly divided along social lines for economic prosperity to deliver one candidate a landslide (assuming a landslide in the popular vote means at least 10% margin of victory.)

In terms of economic growth, I wonder what you mean by "supercharge". I think 4% is pretty unlikely. If the US manages an annual average of 3.0% for the next 4 yea... (read more)

Assuming that 1. the US does in fact hold a nationwide presidential election in 2020, 2. the Democratic & Republican parties get > 90% of the votes in 2020, and 3. US military fatalities in new, unprovoked, foreign wars are minimal (< 3000), I predict: * a Trump landslide with probability 20%, assuming 2% weighted growth over Trump's term * a Trump landslide with probability 70%, assuming 3% weighted growth over Trump's term * a Trump landslide with probability 95%, assuming 4% weighted growth over Trump's term using your definition of landslide, and defining "weighted growth" as annualized growth in quarterly, inflation-adjusted, disposable, personal income per capita, with recent growth weighted more heavily with a discount factor λ of 0.9. (See Hibbs for more details.) This is a clear prediction. I'm more doubtful of chatter about rising political polarization than I am about fundamentals-based models of voting, and those models highlight the economy & war as the factors that most matter. As such I reckon sufficient economic prosperity could in fact produce a landslide for Trump (and virtually any incumbent, really).

I don't think children actually have greater moral status, but harming children or allowing children to be harmed carries more evidence of depraved/dangerous mental state because it goes against the ethic of care we are supposed to naturally feel toward children.

I meant the Chinese public. The actual public of most countries is not all that engaged in the ins and outs of these things.

The Chinese public knows what the Chinese media tells them. It's also very nationalistic.

If true that's mainly an argument against making pointless precommitments you can't possibly enforce. As it happens, I doubt Chinese pay all that much attention to these kinds of diplomatic bugbears.

Why do you doubt it? The fact that they took a US military drone in tit-for-tat, suggests that they do pay attention. If the Chinese wanted it they could leak all those background checks for the US security clearances to Wikileaks or do any number of actions to retaliate more forcefully than just taking a drone as a symbolic action against the symbolic action of a phone call.

The signal sent by Trump is that he will take a call from whomever he wants; the Chinese don't get to dictate with whom he speaks. The idea that it makes China more likely to attack Taiwan is ridiculous.

That doesn't change anything about the fact that it creates strong internal pressure on Chinese decision makers to answer the action in a way that doesn't result in them losing face in front of their citizens.

Another aspect to consider is that Hitler had spent the decades leading up to the war declaring the Soviet Union to be an existential threat and noisily pre-committed to a massive seizure of Eastern European land. He dedicated an entire chapter to the topic in Mein Kampf. So Hitler had perhaps already tied his own hands before he even came to power.

Funnily enough, Stalin totally didn't believe him :-/

I agree, but that isn't what Adams did. Adams first claimed Trump is a master persuader who was virtually certain to win. When Trump was way down in the polls with only weeks left, Adams then switched to predicting a Clinton win, using the Trump controversy du jour as a rationale.

Updating on the evidence would have involved conceding that Trump isn't actually an expert persuader (or conceding that persuasion skills don't actually carry that much weight). In other words, he would have had to admit he was wrong. Instead, he acted like the Trump controversy of the time was something completely shocking and that was the only reason Trump was going to lose.

Adams also frequently hedged his bets and even changed his prediction once the odds for Trump appeared too long to overcome. This is pretty much what you would expect from a charlatan.

I want to be careful in how I talk about Adams. He definitely didn't follow the guidelines for methodological forecasting, such as assigning clear numerical predictions and tracking a Brier (or any chosen) scoring method. As a result I see two main groups of thought on Adams: The first is forecasting oracle. The second is total charlatan (as far as I can tell this is the Rationalist viewpoint, I know SSC took this view). I think the rationalist viewpoint is close to right. If we include the set of all semi-famous people who did/could speculate on an election (including Adams), and then imagine (we don't have the data) that we tracked all their predictions, with the knowledge that after the fact we would forget everyone who was wrong, Adams doesn't seem significantly correct. But if Adams (or an abstracted idea of Adams argument) were correct, It would be because unlike current polling methods it allows for really high-dimensional data to be embedded into the forecast. As of now humans seem to be much better at getting a 'feel' for a movement than computers, because it requires using vast unrelated and unstructured data, which we specifically evolved to do* (I know we don't have great experiments to determine what we did/didn't specifically evolve for, so ignore this point if you want). So, to that extent, current purely model-based election forecasts are at risk of having a severe form of omitted variable bias. As an example, while polls are a little stable, Marine Le Pen is currently at a huge disadvantage: "According to a BVA poll carried out between Oct. 14 and Oct. 19, Le Pen would win between 25 percent and 29 percent of the vote in next April’s first round. If she faces Bordeaux mayor Alain Juppe -- the favorite to win the Republicans primary -- she’d lose the May 7 run-off by more than 30 percentage points. If it’s former President Nicolas Sarkozy, the margin would be 12 points."* And yet has her at ~40%. There is strong prior information
Updating to changed evidence is no sign of a charlatan but behavior of good forecasters.

The idea here is that humanity had started forming true civilizations before 10,000 BC, and a comet impact or airburst over one of the ice sheets caused a huge fireball and flood that led to mass extinctions and the annihilation of civilization

There's no mystery about what caused the quaternary mass extinction--humans reached the Americas and wiped out the ecologically naive megafauna.

Hypothesized to be pushing off the quantum foam.

I'm pretty sure this does not make any sense.

I don't really see how this could be helpful. The biographer would have to be able to discern which qualities made the person successful and translate them into actionable specifics. In practice, it's pretty hard for highly successful people to explain their own success in an actionable way even when they seem to be sincerely trying (e.g. Warren Buffet.)

I've played around with the Playstation VR demo at Best Buy and tried a Galaxy Gear VR at the mall, but was pretty underwhelmed by both. Are the high-end products like Oculus Rift and HTC Vive really that much better?


Yudkowsky showed laughable naivete (or he was just playing dark arts) by citing a bunch of "foreign policy experts" who were against Trump. They were against Trump because they were neocons who might have a spot in a Clinton administration but certainly not in Trump's. (People who describe themselves as "experts" implying impartiality should never be taken at face value--most of the times they are advocates rather than experts.)

Hillary Clinton's state department pushed the "Arab Spring" policies which turned the middle east an... (read more)

Part of the reason I estimated the chance being that high was because I thought (at that time) we were fairly likely to have a recession or major terror attack, which would swing the election to Trump. Neither of those happened, but Trump still won. More recently, II did think the big media company polls were systemically biased by at least a few points in Clinton's favor, so I give myself some credit for that.

I too was wrong. I gave him a 45% chance on this site several months ago and my estimate had hardly changed by yesterday (in fact my estimate got slightly worse, down to 40%.)

Are you really that wrong though, if you gave him a 40-45% chance? Am I making an error to say that based on the real results, someone who was 60%+ sure of a Trump win is more wrong than you are? If I truly believed 538, and that is what I told myself, I shouldn't have been surprised – and yet I was. So what is happening here? I'm not usually surprised by anything that I assign a probability of greater than 10% of happening, why do I feel the way I feel? Perhaps my true probability was <10%.

I gave him a 45% chance

That's actually better than most (all?) pollsters including Nate Sliver.

Agreed, that's pretty much why I think he's crazy. He appears to have devoted an insane amount of effort to a fundamentally pointless activity.

No, if he downvoted all Elo's posts to zero, then it would be a pointless activity.
Not enough data to make a good guess, but sociopathy would be among my top hypotheses.

I don't think so; different types of car are bought by different people and driven differently. E.g. a person who buys a Lamborghini or Ferrari probably likes to drive fast and show off; a person who buys a Volvo probably drives a lot more carefully.

The driver also erred in failing to brake, probably because he was inappropriately relying on the algorithm.

Yep, according to the truck driver, the Model S driver was watching Harry Potter, and it was still playing even after the car came to a stop. He probably had his eyes completely off the road.

The truck pulled in front of the Model S. The Model S had enough time to break and stop but didn't recognize the truck against the brightly lit sky.

What is unclear is whether the driver is likely to have seen it in time if the car had no autonomous mode. Humans, when paying attention even for long periods of time, are still way better at recognizing objects than computers. My expectation is that this is exactly the problematic case for stastical vs concrete risk analysis. The automated system as it is today is generally safer than humans, as it's more predictable and reliable. However, there are individual situations where even a less-statistically-safe system like a human forced to pay attention by having limited automation can avoid an accident that the automated system can't.

What do you think are good ideas for moonshot projects that have not yet been adequately researched or funded?

Leaving aside the bloody obvious things (universal basic income or other form of care, global internet access, etc) Prediction market. They tried but it's dead due to gambling laws. Someone should give it a second try.
Software for automatically playing hypnosis audio via headphones to people who get operated in addition to standard anesthesia.
A friend feed that's like Facebooks friend feed but audio instead of text or images to give people a clear alternative to hearing talk radio.
A pay-for-performance online marketplace for medical services.

I think apples-to-apples comparison is tricky here. Things like the age structure of the population can matter a lot here. A country with an average age of 50 should have a higher level of net worth than one with an average age of 30.

In any case I'm not sure net worth is the valid way to think about "how rich we are" compared to income or consumption or quality of life or whatever.

It would only happen in areas of dense human habitation, which already wrecks the ecosystem. No net harm.

How about disrupting streams?

Quite a while anybody could send information to Facebook users because Facebook implemented the open standard of Email.

Sure, I could send emails to your facebook account, but if I wanted to see any of your social media content, I would have to start a Facebook account and access it via Facebook's walled garden. If I want to use Google+ and you use Facebook....

It's as though you had to use a Verizon phone to have a conversation with other Verizon users, and you couldn't use your Verizon phone to contact people who use AT&T. The outcome is inevitabl... (read more)

Google+ doesn't have a concept of "friend request" or friend. Forcing them by law to do so, reduces the freedom of companies to innovate.
As a rather illustrative case study look at the history of XMPP at Facebook and Google (Talk). Facebook messaging used XMPP until 2015, but it was not federated - but at least you could use client of your own choice. Then they switched to a proprietary API. Google talk used federated XMPP for years, but then dropped server-to-server encryption, effectively cutting off majority of servers, and then dropped Talk in favour of proprietary Hangouts altogether. So the trend is just the opposite - if the player grows big enough and the community becomes self sustained, they will start walling the garden.

I think there is a strong case for breaking up Facebook and Twitter as telecom monopolies. They would be forced to adopt open standards, so anyone could send information to their users, and other companies would be able to create their own clients to send info to facebook/twitter users and vice versa.

Quite a while anybody could send information to Facebook users because Facebook implemented the open standard of Email. I think that feature was mostly used by spammers. I don't want everybody to send me messages in an unfiltered way. As far as breaking up Facebook, I don't see a reason why they should have Instagram and WhatsApp but the core Facebook service can't be easily broken up.

Did he give an explanation of why he thinks that would be a "moral imperative?"

If you are curious, you could've downloaded it off Libgen, or googled the phrase inside Google Books: The context on pg178 is the positive vision of games and entertainment as making lives better and giving people the opportunity to live out their dreams.

I'd estimate Sanders' chances as less than 10%, maybe a bit more than 5%.He would need a mass defection of superdelegates at this point, and it's possible they would be directed to jump en masse to someone else (like Biden) even if the DNC decides to dump Clinton.

Thanks K; good to have more supporting evidence. I won't bother checking out his issues at this time; I'll wait until I know who I get to choose.

I'm not sure that "STEM communities" is a valid reference group for LW.

For anyone curious about this link, I'll save you some time:

From this, they jump to being seriously worried about their inability to control their next Honda Civic because it will have a mind of its own.

It's that type of article.

On a related note: if it is true, does that suggest that, as far as we take the diminishing utility of money for granted, by using extrinsic rewards, we are reducing the number of extreme performers? (in so far as we can't keep giving exponential rewards, and money/tokens/what have you motivates in proportion to their utility).

I think the positional qualities of money compensate for this somewhat. People still work hard because they want to keep ahead of their neighbor/coworker.

Part of my worldview is that progress, innovation and competence in all areas of science, technology, and other aspects of civilization are correlated.

I'm sure they're correlated but not all that tightly.

What would the world look like if that hypothesis were false? Well, we could find a country that is not particularly competent overall, but was very competent and innovative in one specific civilizational subfield. As a random example, imagine it turned out that Egypt actually had the world's best research and technology in the field of microbiology.

... (read more)

Seems like you're just falling for partisan media histrionics and conflating a lot of different things out of context.

On Fox News, Trump said that regarding Muslims in the US, he would do "unthinkable" things, "and certain things will be done that we never thought would happen in this country".

In context, Trump is giving a tough-sounding but vague and non-committal response to questions about whether there should be a digital database of Muslims in the country. He later partially walked this back, saying it was a leading question f... (read more)

Thanks, I guess. I knew he was talking about a digital database, but I was wondering if it could have been a dogwhistle for something else. I don't have a favorable opinion of human decency in general. FWIW, that wasn't a political comment. I hardly ever read or watch anything political. Some TV clips were shown to me by an acquaintance and I wanted an honest assessment of what he had told me it was about. I don't have any opinions on the subject myself.
Load More