When I read this, the Arweave project [1] comes to mind.
Isn't this the exact use case it was built for, preservation of news data in a way that it cannot be tampered with and so that it is kept accessible for everyone forever?
A good security track record must be the most valuable company asset in history.
Apple makes $200B per year from selling the iPhone alone. Plus the services they sell on it, plus deals like the one with Google, plus app store ads, plus cross-selling of other hardware ...
I have one too.
Not because I like the hardware too much. Pixel phones are much nicer, they don't wobble when you put them on a table. Not because I like the software too much. Android is much more to my liking with more freedom to customize it.
But because I have the feeling Apple takes security more seriously.
I wish there was some kind of security arena like there is LLM arena for AI. That gives hard facts about the security track record of phone manufacturers.
It's a bit ironic to think Apple takes security seriously when they infamously delivered ridiculous bugs like 2017 High Sierra root login, which they fixed and then accidentally unfixed again.
What we're seeing is marketing/branding and a genuine for-show effort, all the while they do not audit their code outside some for-show technologies (Siri AI in the cloud).
So the point I am making is that it's all observational bias. You want actual objective security, use GrapheneOS.
Funny since the GrapheneOS devs praise the iPhone for it's security.
I personally prefer seeing real advancement that integrate both hardware and software like their EMTE[1] which no other vendor has had the balls to implement.
And I say that as a GOS user! My next phone will definitely be an iPhone, mostly because I cannot deal with the terrible Tensor SOC. My G4 overheats for nothing! Shameful
The issue is that Android is open while Apple seldom talks about their security issues.
This might make Apple look like the more secure option but the reality may be different because Android is more scrutinised.
If you are sceptical consider these examples: (1) some versions of Apple silicon have unpatchable security defects, (2) Apple at one point decided to not contact up to 500M users affected by a supply chain attack in China due to "language difficulties".
I didn't find about OEMpocalypse from Google talking about their security issues, I found from calif.io. Security researchers don't wait for companies to "talk about their security".
> but the reality may be different because Android is more scrutinised.
This is famously why Linux appears to be less secure than Windows, right? Because Microsoft doesn't talk about their security while Linux is more scrutinized?
> In the simplest of terms, with the bug, if you created a new APFS (Apple File System) encrypted volume on High Sierra, and set anything at all as the password hint, then your password was stored as the hint. In plain text.
I know Apple phones had bugs. Even worse bugs than the one you linked to.
But Google phones had those too.
That's why I said I "feel" like Apple takes security more seriously. And that I wish there were hard facts. Statistics of number of bugs by severity. Independently verified.
At what price, tho? He harms children, he hijacked the attention of billions and helped the rise of facism and hate, which if not addressed will lead into very dark times again, all for a measly 3% out perform?
There's also the allegation that a) Facebook greatly overcounted (on purpose) the impressions from autoplaying videos and b) their ads target more people in an area than physically exist
"Value" is a tough thing to translate. Market cap, and whatever. Considering the fact so many people express difficulty in deleting their accounts - it won't let you, or it gets reactivated later - I'm just going to state the quiet part out loud here; Facebook's valuation is mostly or entirely fraud
If the rise of fascism and hate is due to people being able to talk to each other on Facebook with insufficient censorship, it seems your actual problem is the system of governance that gives everyone a right to vote, even if they do it based on some nonsense they read online.
You are romantizising what Facebook and Instagram have become. The problem is not that people talk to each other. The problem is, that they created algos that optimize for maximal engagement and they get that by triggering the basic human feelings that create the biggest reaction: fear and anger. They literally hacked the human brain chemistry. And they don't stop when they detect that the human is struggling and getting darker and darker thoughts, but they keep pushing to extract even the last second of precision engagement.
"How long and often" people engage with content is a perfectly reasonable metric to optimize and a very close correlate of interest. There are, in fact, plenty of things going on in the world people should get mad about, or that they should worry about. Some people think that's climate change, some people think that's migration, some people think that's rising fascism, some people think that's Mamdami opening grocery stores. The fundamental problem people have is when other people get mad about stuff they like. That's the "fear", "anger", "hate." People who get mad about the things I like are obviously just irrational, emotionally driven puppets of The Algorithm whereas I am Englightened, Rational, and came to my opinions only through Reason and Logic.
What people don't like is huge masses of people coming together and having opinions that counter our modern mainstream elite consensus. Because elites are generally pretty smart that means a lot of these opinions are, frankly, dumb. Some of them simply represent different values (pro vs anti immigration, as one example.) Both are considered intolerable.
It is also perfectly true that people spend lots of time on non-politicized content. Baby videos, cat videos, feel-good human interest stories, all sorts of positive stuff is also massively popular on these platforms. That gets ignored in favor of inveighing for tight control over the political discourse because social media spreads crimethink.
The Western democratic ideal is that you have to convince the masses you're right - a disturbingly large portion of the movers and shakers have decided that instead, we should control what the masses are allowed to say and know, so we don't have to fight those battles.
OK, so, in the vein of the original thread, can you explain how Facebook would be "sowing discord" or "making people depressed and anxious" in some specific way that isn't just spreading other people's speech (ideally, in a way that is based on some objective metrics like engagement) that is Very Bad and shouldn't be allowed?
For example, are you suggesting that Facebook begin censoring information about the climate emergency because it makes people depressed and anxious? Should Facebook make sure that people don't see content they disagree with, because that promotes discord? Or is it the other way around - they shouldn't see content they agree with, because that creates an echo chamber which increases discord when they do encounter other views?
There was an experiment (granted its TikTok, but it's basically the same) where they register a new account, pretending to be a teenager and than start scrolling without liking anything. [1] The amount of extremist content that is shown is really worrying.
To answer your question, yes they should be highly regulated and forced to moderate what they show to users. Maybe even be liable since it is their decision what they show and not the users. And if their business model is then not working anymore, because it's to expensive to moderate the content, then tough luck.
OK, well, I support liberal democracy and all that entails. So we’re kind of fundamentally at odds here and probably won’t be able to find common ground.
That's ok. Thank you for the discussion. But I do wonder, should Russia and China be allowed to use these tools to sow distrust against our Democracy? For me the tolerance paradox formulated by Karl Popper comes to mind and it is a clear no.
Why is it this line of reasoning never goes into questioning just who the forces endlessly propagandizing about the need to censor social media are? Is this not sowing distrust in democracy itself? (If you can’t trust your electorate not to believe Chinese propaganda, what does that say about your attitude towards democracy as a system?) Who is doing this and why? That would seem to be very salient since much of the reporting on bot campaigns has since been discredited, with whole organizations confabulating these stories then laundering them through the media to push their desired political outcome. (C.f “new knowledge.”)
Foreign propaganda in democracies is as old as democracies themselves. It is by no means anything new; foreign influence campaigns spent enormous sums and labor trying to sway democracies from the very beginning - often very successfully. I don’t see why it’s any different for it to be published on Facebook instead of laundered through print media and rumor mills. At the end of the day you either recognize these problems and appreciate that trusting your electorate/having a liberal democracy is the best option we have compared with the alternatives…or you don’t.
> At the end of the day you either recognize these problems and appreciate that trusting your electorate/having a liberal democracy is the best option we have compared with the alternatives…or you don’t.
Many of those democracies are trending illiberal thanks to the useless yet arrogant liberal decrying of “there’s simply nothing else we can do” and championing a rotting status quo.
They are trending illiberal because of the tension between elites who think ignoring and censoring the popular will on various issues is wise and the voters who see that and think the only solution is a drastic one.
Several western countries could kill their more extreme parties overnight simply by enacting a few reforms with immigration and/or affordability.
Democracy does not imply that people should be allowed intentionally make people depressed and anxious for profit. There's no reason that must be allowed for a healthy democracy to work.
Liberal democracy does not mean that the government and other authorities hide information and opinions from people for fear it might make them “anxious and depressed.” I gave you a specific example: should they stop letting people see content relating to the climate emergency because it makes many people anxious and depressed?
That is the view of paternalistic authoritarian states whose primary concern is public harmony, not liberalism.
We're not talking about the government and authorities, we're talking about Facebook. The government showing people information that makes them anxious and depressed isn't the same as a for-profit entity intentionally tailoring individual feeds to maximize people's specific anxieties and depression because they discovered it generates billions of dollars, allowing their executives to live in opulent luxury floating on mega yachts. That whole thing isn't a part of liberal democracy.
OK, so who is deciding what information makes people “anxious and depressed” and therefore what Facebook is allowed to republish in feeds? Would it be reasonable for Facebook to censor all information about the climate emergency to protect people from feelings of anxiety and depression, even if they engage with it?
Content that speaks the truth, like content about the climate emergency is not the problem, even if people don't want that to be true. The content that are just blatant lies, nowadays even published by elected officials, that they just publish because it is in their own personal and financial interest is.
Right. So instead, a central authority determines what is true or not and therefore what is allowed to be published. Easy - there’s clearly a neutral arbiter of truth we can all depend on! So the problem was never that the algorithm let people see content that was depressing and anxiety-inducing - that’s okay - but when it does so and it’s not true. How the goalposts move.
You’re right back to what authoritarian states say and do, and diametrically opposed to what underpins liberal democracy: the idea that the demos decides what is true and good. That idea has lots of very obvious problems but we decided it was preferable to the other systems, one of which you are now advocating for. In a liberal democracy we all are responsible for persuading the people that lies are lies, not censoring them.
If you believe that the Algorithm is just too powerful and you can’t compete in the marketplace of ideas against people lying - well, you don’t believe in liberal democracy. You believe in an ideologically authoritarian state that tells you what’s true or not, and where people are not allowed to make free, (un)informed decisions.
That price is paid by us, not Zuck and his shareholders, so it was at no cost to him. Such is the problem with capitalism. Actually the rise of fascism has been great for Zuck and his shareholders, they'd like to see more of it.
The dangerous situation we are in is that entities outside of the USA hold significantly more US assets than US entities own assets outside the USA.
So if the USA declares "ownership be damned", it would be a net win for the USA.
Or in simpler terms: The USA has been giving out IOUs for real goods for decades. It is now tempting to default on those IOUs. The USA did this before. Up until 1971, a dollar certified a claim to gold. Then the USA simply canceled that promise.
The current ramp up to AI dominance might be what holds the USA back from breaking more ties with the world. It is a race that must be won. And investments from abroad help. But when the whole world is operated by US AI factories things might look different. When every task around the world is performed by intelligence generated in the USA, just like every computer is run by US software now - there might be little incentive left to "pay the rent" for the rest of the world that now sits around idle, fiddling their thumbs.
> The dangerous situation we are in is that entities outside of the USA hold way more US assets than US entities own assets outside the USA.
The dangerous situation we are in is that the US is berating and betraying its closest historical allies, eroding its credibility and soft power, and weakening its leverage at a time when threats to the Western order that has existed since the end of WW2 are growing rapidly.
Incidentally, the orange man in charge loves to tout "deals" in which foreign countries agree to invest substantial sums in the US, so any actions that lead foreign countries to believe their investments might be nationalized banana republic-style won't help the administration's efforts to attract investment and onshoring.
I would guess that most of those announced deals will never come to fruition. He loves to pat himself on the back, and everybody seems willing to go along with it, until they actually have to spend money.
Given the last 19 months of tariff chaos while Republicans control all branches of government, I don't see any reason to think they're concerned about that.
Plus the whole "assassinate the Iranian president in the middle of negotiations to start a war across a major global trade-route" thing.
> So if the USA declares "ownership be damned", it would be a net win for the USA.
This opinion is very naive. The moment the US declares "ownership be damned", the US dolar stops being the currency used in things like the oil market and it's value tanks. The world will scramble to use an alternative currency, which will end up being either the euro or the yuan. This effectively marks the end of US as a world power.
> Or Bitcoin. Of which almost half is already under US control.
This sort of opinion is laughably out of touch. Listen to yourself: oil markets will move away from the US dolar due to how the US governments destroying its value, and instead will move to bitcoin because... The US controls it?
Are you aware you are posting these outlandish opinions on how the world trusts the US with it's currency in a discussion where literally European countries are pulling their gold reserves out of the US because they no longer trust the US?
It's not going to be a net win for anyone. A country that needs to keep borrowing to stay afloat is not going to improve its position by effectively defaulting on its debt, regardless of the exact method or type of assets involved.
The USA might hold a majority of the worlds value generation in a few decades. If AI generated in the USA powers everything. Then the USA does not need to borrow anymore.
That outcome seems unlikely. Both that AI will remain dominated by USA for a few decades, and that they will use newfound profits to eliminate budget deficit.
And why would they want to default on their debts in this hypothetical future where they've "made it", raking in AI profits from everyone. They would work hard to preserve that world order and not shake things up unnecessarily.
The USA has managed to dominate software for decades now. Why would it change? Take a look at the current datacenter investments in the USA vs other countries. It looks very much like the USA will even extend its lead.
> why would they want to default on their debts
So they don't have to share the AI profits with foreigners who hold shares in the AI companies.
> So they don't have to share the AI profits with foreigners who hold shares in the AI companies.
They can do it now. They don't, because it's not worth it. It won't be worth it in the future either. They'll have yet more to lose. Especially in this hypothetical scenario where their massive AI exports are what's paying for their economy – good luck getting your counter-parties to pay for your exports after you have decimated their economies and soured relationships over a one-time boost that will cover what, a couple years of US government spending at best?
> The USA has managed to dominate software for decades now. Why would it change? Take a look at the current datacenter investments in the USA vs other countries. It looks very much like the USA will even extend its lead.
- Because USA managed to dominate other industries for decades as well, only to lose them for one reason or another?
- Because China has shown its ability to dominate in other complex modern industries like EVs and drones... and software is not special, and they're already starting to out-compete US with cheaper open-weight AI models?
- Because trust in America has already shaken, and work to get off US software in other countries has already started, even if it's slow and will take a long time?
- Because the contents of datacenters have a useful life of only a few years, and over-spending hundreds of billions on first-generation inefficient datacenters may sooner result in a financial crash wiping out most of the industry, than build a meaningful long term moat?
- Because US doesn't even produce the chips that power said datacenters, whereas China, their main geopolitical competitor, is working full time to develop their own chips, while militarily posturing against the island next door where all America-bound chips are made, at a time when US ability and will to do anything about that seem to be at an all-time low?
- Because after enough iterations, the AI race may eventually boil down to in large part to competition on the cost of electricity, in which US is at a disadvantage?
US is not going to willingly default on its debt or violate investors' property rights. That's just a fantasy that got nothing going for it.
> good luck getting your counter-parties to pay for your exports
It's not a matter of choice. Once a country runs on US artificial intelligence, it can't just say "Well, let's do things manually again". People won't know how to do it anymore. There will be no more keyboards, steering wheels or cash registers to operate.
It's true that China might offer an alternative. But changing a country from being run by US AI to being run by Chinese AI is not a small feat and comes with its own issues.
> US doesn't even produce the chips
TSMC is running and expanding a factory in Arizona. Budget: $265 billion. Intel, Samsung and Miron are also building capacity in the USA.
yes. but most of it is not to other countries. Most government debt (over 80% globally) is privately owned through sovereign debt instruments like bonds.
The examples they show are someone typing something into a prompt and then some agent starts coding.
But the hard part in automation these days is not writing software. It is "how does the software get the credentials to access data and services? What APIs does it use?". So before being able to type a prompt and then use the software created, there must be the hard part: Setting up all those interconnectivity pieces. Making sure the software can access and change data, trigger actions, but only within a certain scope.
It will be interesting to see if SAP can turn into the basis for companies that use AI heavily or if companies that use AI heavily will prefer to replace SAP.
Who are the contenders that start fresh with focus on AI?
One that comes to mind is IONOS. They seem to be in the process of building something like a "SAP in the times of AI" from the ground up. Starting with the demands of SMEs. It looks like an interesting approach to me: Having a "knowledge base" for each customer, initially filled with the data from the website. Then let the customers add more data. Then offer agents that do work for the customers. Answer calls and emails and do marketing for example.
I have in mind what I think is a very strong framework to allow companies to replace legacy ERPs quickly with AI help. I guess I should get on with it and publish something!
> It will be interesting to see if SAP can turn into the basis for companies that use AI heavily or if companies that use AI heavily will prefer to replace SAP.
Considering the human and institutional complexity of implementing ERP, CRM, whatever the hell Concur is supposed to be, etc. I'd bet on most organisations implementing AI around and with those systems instead of replacing them with something vibecoded / "AI native" vendors.
Emarketer’s data finds that standalone chatbots like ChatGPT,
Microsoft Copilot app, Google AI Mode, and Amazon Alexa for
Shopping (formerly Rufus) in U.S, will generate less than $1
billion in ad revenue this year, and just $5.41 billion by 2030.
How would that be possible? In the past I used Google maybe 10 times a day with a short query. From which Google had to guess my intent. Now I babble with Gemini all day about everything. And Gemini can ask questions what exactly I mean. Why wouldn't Alphabet be able to generate more revenue from this than from search? And Google's ad revenue from search is over $100B per year.
Just because there are no ads now does not mean there never will be. Google search was run without ads for the first years too.
> How would that be possible? In the past I used Google maybe 10 times a day with a short query. From which Google had to guess my intent. Now I babble with Gemini all day about everything. And Gemini can ask questions what exactly I mean. Why wouldn't Alphabet be able to generate more revenue from this than from search?
When you used google 10x a day, you saw, at a minimum, 20 ads/day. How many ads are you seeing when you babble all day long with a chatbot?
> When you used google 10x a day, you saw, at a minimum, 20 ads/day. How many ads are you seeing when you babble all day long with a chatbot?
There is also way more competition between chatbots than there was between search engines after Google started leaning into ads. Which strengthens the Khan-Esayas hypothesisis "that data privacy constitutes a key parameter of non-price competition in the market for" consumer tech [1][2].
why would they be able to generate more revenue? it isn't clear to me. Maybe ads will be targeted better, maybe they wont. Either way, major companies marketing budgets won't be going up, only shuffling around?
So its not really much to do with the technology or how good it is, but the stories their salespeople can tell. And I guess the story isnt good/convincing enough
The same kind of speculation questioned whether Facebook would ever have a significant business. Before that, the same was asked about Google's business.
If GPT can maintain or grow usage, the ad dollars will be there given the enormous scale. There is a hundred billion dollars plus worth of advertising waiting in the LLM space. It would be surprising if Facebook doesn't contract for example, losing ground to LLMs on advertising over the coming decade.
They have to build out all the backend ad tech crap that google and fb have built over decades. As in the campaign tools, scheduling, targeting, analytics, auctions, exchanges etc. Advertisers and publishers have already deeply integtated into that eco system. Also the google and fb system has created a content generating army that is incentivized to fight tooth and nail for eyeballs. OpenAI doesnt reduce the activities of the eyeball capture army, it just freaks them out and pushes them into overdrive.
WhatsApp and Messenger have billions of users but don’t make much ad revenue. I’m sure there exists a way to show ads in ChatGPT that will be figured out eventually but so far nobody’s figured out how to monetize chat as a medium
Also it’s basically free for Google to show you a sponsored result but embedding one in a ChatGPT response actually costs money (assuming they’re part of the generated response).
Lastly I will bet you one Stargate datacenter that Meta has thought about LLM-based advertising, and if there’s any low hanging fruit there it’s already been tried
> WhatsApp and Messenger have billions of users but don’t make much ad revenue.
Not directly, but I bet that connection graph is worth millions. Same with the Giphy keyboard purchase. There's a whole unseen economy of data being valuable for reasons that aren't widely discussed.
Facebook was in financial loss until it got really serious about pushing ads and outrage. People who speculated that Facebook is not earning much on ads while Facebook was not earning much on ads were correct. Same with Googles business.
And also, very relatedly, both companies enshittified their products great deal in the process. Google search used to be better for average user, Facebook used to be better for average user.
Genimi can start pushing ads everywhere, can start using dark patterns to better manipulate us and one they will. They will start earning more money at that point.
In theory, proof of work that is used to mine a cryptocurrency could be a solution.
Bitcoin and others are already secured via massive pow computations. If we could shift that into browsers, no additional energy would be used and we could solve an issue that has been unsolved for too long: How to pay websites that provide useful information other than with ads.
The question is which resources typical consumer hardware has that large centralized compute power does not. In-browser POW to pay websites would only be possible if such a resource exists.
I am not familiar with the topic, but maybe CPU power and memory? Both seem significant in a typical consumer device.
Napkin math: If a consumer device can generate $100 per month, that would be 100/30/24/60/60=$0.00004 per second. If the user waits for 5 seconds before the first pageview, that would then make the website provider $0.0002 per visitor. Serving a million visitors per month is nowadays easily possible on a $10/month machine. So the $0.0002x1000000 = $200 would make the website a nice profit.
Proof of work, even "custom", where the user does not need a particular interaction with the page, does not work. The scrapers are running headless Chrome and solving the work. They do not care, they do not pay the bill, the compromised system's owner pays the bill.
I have such system for the registration form on one of my website to prevent the double validation of emails to be used to spam emails of victims. The PoW challenge prevents less than 10% of the bots.
First of all I am very critical of the $100 per consumer device per month figure.
Also all major browsers block crypto miners on webpages now (for good reasons) so it may prove difficult to allow "good" mining scripts while still blocking "bad" ones.
I could’ve said a lot of things. Kiwiflare is the only one I’m aware of (and can remember name of) that provides a visible proof of work occurring in the browser, regardless of reputation scoring (which I don’t think they use). Although, I suppose by definition, anyone using that site is of poor character…
If you can think of another example, please be helpful and provide it, thanks.
Better skip the PoW part as it'll be wildly inefficient for most of the work done.
Instead, exchange web traffic for actual $. Say, some kind of tokens that are easily turned back into hard cash through a 3rd party.
Requesting a 100KB file? Okay, that'll be a $0.00002 token, please! (visitor's user agent provides it in a manner transparent to regular web users). Requesting a 3MB image? Okay, that'll be a $0.0005 token, please!
Result: niche websites earn hard cash. It doesn't matter much if you're hammered as long as the hammering comes with a corresponding flow of tokens (read: $). No token(s)? No service.
Regular web users would pay for those tokens through their normal internet service fees, and otherwise not be bothered. Massive scrapers would have to pay somehow for the tokens to be served web data at all.
In effect: put the bulk of public web sites behind a paywall. But with the bar low enough & in a manner that it's transparent for regular web users. Clicked "reload" by accident? Oops, internet service bill got upped by 2 micro-$.
> partly because it causes annoying delays for those trying to get to the site
This is true but usually a small issue. It’s further alleviated by cached tokens so you only have to solve the challenge once in a while per site, and a login token may let you skip it.
> partly because it seems inevitable that the scrapers will eventually find their way around it…A proof-of-work requirement is not a huge obstacle when you have millions of other people's machines to do the work on.
Isn't this the exact use case it was built for, preservation of news data in a way that it cannot be tampered with and so that it is kept accessible for everyone forever?
[1] https://arweave.org/