I’m kind of amazed when I read comments like this, but I have to remind myself that I work in an industry which use these tools at the cutting edge and see what they can really do. In the space of 18 months I changed from skeptic to the belief that our world is going to RAPIDLY change, and soon.
I sense that statistics and benchmarks and research and statements from the world’s greatest academics won’t sway you, so maybe I’ll give you a personal anecdote. I have suffered from a condition my whole life called bile acid malabsorption. It caused chronic diarrhoea, pain, arthritis, dehydration, insomnia, and more. I spent decades searching for an answer. Dozens of different tests. Eventually doctors just said I was depressed and prescribed me antidepressants. They didn’t help. On the bad days I considered ending my life.
In desperation I turned to ChatGPT. Over months I described my symptoms, triggers, diet, timing, etc. We “sparred” with each other over assumptions and ideas. I gave it all my medical history. All the tests. Eventually it concluded that BAM was likely (plus another few options). So I pushed my doctor for a specialist referral. The specialist agreed to a scan based on the symptoms. It was confirmed. I’ve been taking some cheap medication each day now and it has changed my life.
I know others for whom ChatGPT has changed their lives in similar ways. Research shows LLMs are better than doctors already in many cases at diagnosis. They are improving at an exponential rate.
I have no doubt there are similar stories to yours or to a lesser extent. And while I’m glad you have gotten some relief, what price are we paying as a society? I’m not just talking about the environmental impact alone, what about the societal impact? Social media has not been around too long but it seems to be a net negative for society. I’m sure there are plenty of anecdotes to the contrary, but the studies are showing more and more of the detrimental impact to kids/teens. The stories and testimonies of how it was engineered to be addictive. Yet we’ve mostly moved on and allowed money more social media to get owned (and now traditional media) and controlled by the rich. But your statement is right, I doubt I’ll get swayed easily. Anecdotes and proclamations work when thinking short term. But taking the long view or looking at recent history I don’t see a positive. We have the hype being brought to us by the people that gave us the dot com bubble, crypto and NFTs. But hey, go fast, break things and never ever think about if what you are doing is really beneficial to society (after us, they are not us.)
I think most of the nuance in your first argument was undermined and overshadowed by your fairly categorical assertion that LLMs are bullshit generators and not actually useful.
When you got a reply illustrating how incorrect that claim from your first argument was, you shifted to focusing on the other argument (the one I actually happen to agree with - the cost to society of hitching increasing dependency on big tech will make the social media harms look like childs play).
I think your argument will be better received if you focus on the very valid concerns of societal harms, and acknowledge the ways LLMs are tremendously capable, without downplaying that.
I'm with the person you replied to in seeing how capable LLMs can be when you spar with them appropriately. They confabulate, but that's your job to catch as a sparring partner. But they do bring useful knowledge of thousands of PhDs into conversations - and even if you're among the most erudite humans on the planet, this is still an asset in intellectual search for truth on many topics.
Back to the genuine problems, and they are many: this power, concentrated in the hands of big tech, is a multiplier on the power already concentrated there, with many new capabilities - especially scary being the capabilities for subtly influencing and manipulating both individual and group behavior - for profit or otherwise - by the companies or their customers, or governments, or... The possibility space of harm and abuse is large..
On net, I think we should all be pushing to educate everyone around us on the pros, the nuances, the risks, and the big cons, and working to try to build a future of offline models rather than subscription service dependency..
Bonds are no longer recommended. Current research indicates 100% equities to be the best composition leading up to, and past, retirement.
To point, the economic uncertainties around geopolitics, AI, and war, plus irresponsible debt spending by governments and the prospect of QE (and higher inflation), is pushing long term rates steadily higher. There’s a reasonable chance that 30y treasuries are nearing 6% by the end of next year. Remember that rates and bond prices are inversely related. Anyone who holds bonds in this market will likely lose money. Holding to maturity won’t help much either because if inflation continues to rise, as is a major concern, most or all of that 5% yield gets eaten.
> Bonds are no longer recommended. Current research indicates 100% equities to be the best composition leading up to, and past, retirement.
Are you referring to Anarkulova et al? Might be worth mentioning that the fixed income part is replaced with international equity, not more domestic equity.
Their definition of long run and your definition of long run are probably different.
Also, it should be noted, just because it's the optimal to have the most $'s that shouldn't be the goal. The goal should be to survive your retirement with "enough".
And it should also be mentioned, most people can't stomach holding 100% equities, for a very good reason. When the 40-60% market crash happens, people get emotional and make emotional decisions. Sure there are the lucky few that can hold out, but most can't. Are you going to be one of the few lucky ones? If you haven't yet been through it once(last one in the USA was 2008/9), how do you know for sure?
That’s been something I’ve started doing. The nice part of the bond chunk of my investment portfolio is the current income aspect of it, with monthly dividends that give an annualized return of a touch under 4% on top of the capital growth.
For most people, $VT (or VWRA) is optimal. You should have a U.S. tilt because most growth is coming out of the U.S. $VT will naturally rebalance into international equities on that growth. If you already have a U.S. heavy portfolio and want more international exposure, $VXUS.
Bonds only give you certainty to the extent that inflation remains certain.
Stocks generally rise with inflation, whereas bonds continue paying out the same nominal amount, which buys you less over time.
As a retiree I'm 50/45/5 in stocks/bonds/cash, having opted for a conservative portfolio. The stocks are the only reason I haven't lost buying power. But the bonds have performed so poorly that I've barely kept up with inflation despite the amazing bull run in stocks.
Are we talking about bonds or government bonds here? The former will beat inflations assuming you don't just buy AAA rated ones. Investment grade perpetual bonds in US dollars yield over 6.5% on a Yield-to-call basis.
It depends on the goal / priority. In most financial / retirement advice they are focused on average middle class Americans. They tend to have too little savings, and not a lot of options.
If you have more than enough saved to meet your basic needs, it does (IMO) make sense to give up some total income for lower variance.
I sleep on certainty. I feel bad for the people based their futures entirely on a trajectory from a time we'll look back on as "utterly unsustainable".
Bonds will give you poor (probably negative) real returns, but if you're 10-20 years away from dying you're more concerned with wealth preservation than growing your wealth.
People have forgotten this but equities are an infinite duration asset that are prone to periodic, significant, often violent crashes.
(Edit: often at a time when everyone is absolutely convinced they're the best asset class...)
You can keep some equity exposure but you don't want 1929 or 2008 to happen the day after you retire when you might live for another 30 years
> SpaceX is valued at a whopping 94x revenue. This deal increases SpaceX's revenue by $11 billion per year. If SpaceX maintains this revenue multiplier, then this single deal boosts SpaceX's valuation by 94 x 11 billion = $1 trillion dollars.
This isn’t how valuations work. The PE ratio isn’t fixed. It doesn’t scale with revenue. It’s based on projected future growth. This kind of deal is expected, meaning this deal likely won’t move SpaceX’s market cap much. Certainly not by anywhere close to $1T. That’s +60% of the entire pre-IPO market cap.
Google is doing this because they need more compute and TSMC is booked out for years.
Given the incredible progress of local models, on present trajectory I think we see comparable levels of performance to frontier models in two years on 128GB unified RAM and 6-bit quantisation. Note how the frontier models are now hitting superior benchmarks with only 200,000 tokens. I think we still have a long way to go with distillation.
System RAM has much lower bandwidth and less predictable access. Notably, the transfer from system to GPU is very slow. About 30x slower. LLMs aren’t designed to queue or parallelise operations to account for this. They just become much slower.
We don't want anti-capitalist reforms. Perhaps reforms like those we use in Denmark would make sense? We're ranked higher than the U.S. in ease of business. We love capitalism here. But we also have high taxes, universal healthcare, and great social safety nets. It allows us to lean into the great parts of capitalism while also protecting those for whom capitalism isn't working so well. It's a great compromise without having to go full starvation and gulags, which would be worse than what we have now.
In plenty of places in the US we have high taxes and none of those things. Since it's HN I'll mention San Jose has some of the hardest water I've seen in any city and I've seen glass on the street for 6 mo before being cleaned up. I can't figure out where the taxes are actually going. Other states I've lived in I've even paid less and gotten way more.
I don't think people mind paying taxes, but it's when your taxes don't clearly benefit you that people get upset.
Honestly it seems to me like one party just wants to shift all costs to the poor (rather than the government) and the other wants to be the king of Nottingham. It's no surprise our citizens are feeling defeated. The choices appear to constantly be the lesser of two (geriatric) evils. Not choices between reasonable leaders, but with different beliefs
It really does seem like the U.S. gets the worst of both worlds: high taxes (well, not quite as high as Europe), and poor social services. Even worse: despite all this tax revenue and poor social services, the nation is borrowing trillions more every year. Where is it all going?? I suspect there is a lot of fraud baked into the structure of the system, and this makes the governance layer very resistant to change. I've seen many attempts to move to proportional representation over the decades and both major parties rise up to quash such attempts with fury.
I hope you guys find a better way forward. I have affinity for your people and culture and I think most of you have big hearts and mean well.
I don't know what starvation and gulags have to do with anti-capitalist reform. In this country (the USA), capitalism is what produces gulags and starvation.
On the contrary. The U.S. has so much food that 72% of Americans are fat. There is simply too much tasty food available. This happened in a very short space of time, historically speaking, thanks to capitalism. As for gulags, one would need to use a very cynical definition to believe that. Especially when we have real gulags around the world today.
I imply that the alternative to capitalism is starvation an gulags because that is roughly what happened the last 37 or so times humanity tried something else. That's just from the last century. Capitalism isn't perfect, but it's infinitely better than everything else.
Let us hope this results in farms offering Americans better wages for working on farms. Only 1% of the retail cost of produce is due to labour. Farms could double wages to workers and it would barely move the needle on inflation.
I agree, but the farmers aren't pulling in massive amounts of cash either. There are record profits being siphoned from the wages of the workers, and from the profits of the farmers, and from the pockets of the consumer at the register. We'd all be better off if we tracked the money down and took it back.
Put Wall Street out of work and have the bankers work the farms out there in rural Conneticut and NJ and NY. It'll be good for them, put some hair on their chest and give them Real World Experience (tm)
I agree. For what it's worth, I think Google's business model is in trouble. They're putting in heroic effort to stay relevant during this shift to AI search, but how does their advertising model work when people expect summarised and accurate information instead of scrolling pages of ads?
The search space is actually not that unhealthy. There are competitors. They just lack the reach of Google. These competitor will become increasingly useful when our personal AIs plug into them to search for the right info.
That said, I think the old internet business model is dying. Why will people write articles and pay for hosting if they can't receive any advertising revenue? I think I'm okay with this. Before advertising decimated the free internet, people ran sites for fun. Maybe we return to that.
>> but how does their advertising model work when people expect summarised and accurate information instead of scrolling pages of ads?
1) because that fancy summarised results page will effectively be the ad. The results will push you towards google's partner sites or promote behavior benificial to them.
2) because partners may soon be able to pay to change the results of searches. Who won an election 20 years ago? That "fact" from the AI could depend on which answer has been sponsored by which political interest. What happened in china in 1989 ... how much money do you have?
3) the user becomes the product. No ads in search results, but our searches and online behavior tracked by them will be sold to whoever is most interested in what we do online.
TL;DR: Google is heavily stuffing more and more ads into every monetizable SERP, which explains the ever-increasing revenue.
But my theory (expanded in more detail here https://news.ycombinator.com/item?id=47957708) is that its cash cow is heavily optimized towards its current anti-competitive, multi-sided chokehold of the online ad market -- look up the AdTech Antitrust findings around manipulating ad auctions, e.g. Project Bernanke -- which relies on ad volume.
With agents however, I suspect that volume will shrink drastically even if they stuff ads at each step of the conversation. Because it's only the final click that will matter, and those ads will be meaningless.
It's not clear that Google can charge enough for that final click to compensate for the loss of both, the value of all the ads that lead to it, as well as their ad auction manipulation premium.
Hanging my comment here, not blaming the parent poster, it's just relevant...
All the "AI search" I've tried is far less relevant than a proper search, using search engines without it seems much faster and more relevant.
Whenever I get AI search results, its summaries lack substance, are often just plain wrong, and so on. When I check, it's often usong the wrong aggregate data to return results.
Search engines these days, traditional ones, have been hobbled by targeting the lowest common denominator. They drop search terms, alias them (eg, Joe=Joseph,Josephene, Jodiene), and thus return endless bad results.
Using "web" search on Google, for example, putting all search terms in quotes, helps enormously. If searching for Joe Baker, you don't want aliased responses.
My point is, AI is yet another abstracted layer. It not only aliases things, it halucinates, and further, it doesn't even show the context it simply summarizes.
It's the next layer of crappiness.
As a side note, there os a
product in my home country called the Swiffer. It is like a broom, but uses disposable sticky paper instead of broom bristles and a scoop.
Point is, it isn't very environmentally friendly. I weirdly see environmentalists using it, instead of just a good old broom.
And now I see environmentalists using AI for search. What? Why! The sheer additional power and cost, all the new datacenters, hardware, just for what??
Even of one believes it is mildly better (which it isn't, if you use traditional search properly), are you now replacing your broom with a swiffer?!
The amount of environmentalism that goes out the window, for convenience, astonishes me.
It is totally dependant on context. Ask about product purchases/prices and it will give you great results. Ask for basic data and it will quote wikipedia or government websites all day. Ask for a coding solution and it will spit out answers ripped from the corners of github. But i asked google about hydrogen-3 recently and got back a result saying that we were mining it on the moon. An entire space economy was invented because AI cannot tell scifi from reality.
The real test: dont test AI by asking it questions you already know. Ask it a proper question and then go out and find the answer yourself. That is where AI really falls down, the non-trivial stuff.
I’ve experienced the opposite. I get good summaries with links for original sources when something seeems off, which it almost never does. I also like that I can glance to see where the data came from.
Several months ago, I was getting hallucinations and silly answers, like Jaco Pastoreous being the bass player for Metallica.
but lately it’s been great. Especially for technical syntax stuff, like asking for syntax on a jq command, or sal syntax, or something trivial like movie casts members….
Edit: I also assume there’s now some pre-processing filter to retrieve answers from a cache or something, because I’m getting pretty long answers very quickly….
I've noticed something of a hybrid of what you posted and what one of your responders below posted.
For average results, that can have an acceptable answer that tends to the mean or societal mainstream views along some subject, the google answer works for me. For others, which are well unfortunately most of my queries these days, the google answer is brainwashing and propaganda. It feels little more than a continuation of the SARS-COV-2 era fact checkers which were actually failed attempts at propaganda machines for enforcing certain views as true while others were cited as "evidence does not suggest" (hand selecting the evidence of course means this is nothing more than confirmation bias).
I think the AI technology used in most LLM models is an acceleration of this- it will tout mainstream views or hand-picked views as truth while calling others false. And without an ability to search any more for opposing views, you get left with a really warped view of reality that might work for most. Though not for everyone.
>All the "AI search" I've tried is far less relevant than a proper search, using search engines without it seems much faster and more relevant.
I've had the opposite experience with Perplexity Pro. Yesterday it took just under nine minutes to deliver a detailed and accurate answer as to the nearest locations where I can buy Meta Ray-Ban Display glasses (at least 3 hours drive away FWIW) that day.
In the past several months I've spent countless hours using search engines with dismal results that weren't close to that.
Regarding a swiffer: it's a zero sum game. Where did the materials come from? The earth. Where did they return after use? The earth. The only thing changed was the energy used to create and transport them.
If conserving energy was important, none of these AI datacenters would be built, none of our items would be shipped across the planet from Asia. They would conserve energy and import raw materials only.
While I agree in principle, I think it was inevitable. People been gaming SEO for so long that even with judicious use of search operators, it was getting harder to find the things I wanted (probably drowned in a sea of spam in such a way as to fall outside of the optimized search index). It _looks like_ the AI overview does not have this problem (yet...).
Anecdotally, I find Google's AI Search results to be genuinely helpful much of the time, and the box isn't so intrusive that it prevents me from digging into literal search results if I am not satisfied with the AI summary.
The new business model is misinformation sold by the highest bidder. Ask it what's the most reliable kind of computer, it answers with whoever paid the most this week.
> Which then begs the question, what is IQ actually measuring - something more like innate intelligence, or a fairly big slice of learned, habituated test-taking ability?
> "It is a variable that summarizes positive correlations among different cognitive tasks, reflecting the assertion that an individual's performance on one type of cognitive task tends to be comparable to that person's performance on other kinds of cognitive tasks."
In other words people who are good at these tests are also good at real world tasks. Meaning IQ measures much more than one's ability to pass an IQ test. There are of course many examples of poorly structured IQ tests (including people re-taking the same IQ test and doing better at it the second time around). However a well-structured IQ test presenting novel questions (absent popular culture references and trivia) provides a very good approximation of the g factor in almost all cases. This means high IQ is highly correlated with things like income, unemployment, crime, homelessness, addiction, divorce, and many other objectively measurable life outcomes.
There is room for a philosophical debate about what g factor is, but it is beyond contest at this stage that g factor is real, and IQ almost always does a very good job of measuring it.
It's beyond contest that the g factor is real in part because it's a statistical inevitability in any series of related tests, be they for intelligence or product/market fit in automobiles. It's an exploratory statistic, a hint at underlying causality; it is not a dispositive revelation about the structure of human thought.
Sure: there's a battery of general cognitive tests, and if you smush data sets together a dominant factor will emerge. And?
This is exactly my point. It's tautological to argue that IQ isn't correlated with intelligence. It is, definitionally. You seem to acknowledge this so perhaps I don't understand your original comment.
(I misread your original comment --- no, it's not tautological. It's tautological to say that IQ isn't correlated with IQ test results. "g" has nothing to do with "intelligence" as a construct.)
If IQ tests measure one's ability to do most other cognitively demanding tasks then it's measuring more than the test itself. Whatever you want to call that thing is irrelevant. In sociology and psychology we call that the g factor. The vast majority of people call that intelligence.
That's the tautological argument! Note that I'm not even really engaging with the validity of IQ tests, just the surface logic that you're using. You literally just argued, in effect, that IQ tests measure intelligence because people call them intelligence tests!
That's a semantic point you're making, and I'm not contesting it. To repeat myself, you can call the correlation anything you like. I'm merely explaining that the correlation exists and is incredibly material.
We agree the correlation exists. The incredible materiality of it is what's in dispute. The obvious problem with these arguments is that people point to the correlation and say "see, I'm right about the materiality!".
We have strong data to show that people who score higher on IQ tests also earn more, have higher net wealth, have lower rates of unemployment, higher graduation rates, lower suicide rates, lower rates of addiction, lower rates of crime, lower rates of divorce, lower rates of fatherlessness, lower rates of household abuse, higher rates of home ownership, longer life spans, have better general health on almost every metric, and a host of other positive QoL indicators. I don't see how you could argue that is not material.
I sense that statistics and benchmarks and research and statements from the world’s greatest academics won’t sway you, so maybe I’ll give you a personal anecdote. I have suffered from a condition my whole life called bile acid malabsorption. It caused chronic diarrhoea, pain, arthritis, dehydration, insomnia, and more. I spent decades searching for an answer. Dozens of different tests. Eventually doctors just said I was depressed and prescribed me antidepressants. They didn’t help. On the bad days I considered ending my life.
In desperation I turned to ChatGPT. Over months I described my symptoms, triggers, diet, timing, etc. We “sparred” with each other over assumptions and ideas. I gave it all my medical history. All the tests. Eventually it concluded that BAM was likely (plus another few options). So I pushed my doctor for a specialist referral. The specialist agreed to a scan based on the symptoms. It was confirmed. I’ve been taking some cheap medication each day now and it has changed my life.
I know others for whom ChatGPT has changed their lives in similar ways. Research shows LLMs are better than doctors already in many cases at diagnosis. They are improving at an exponential rate.
reply