Gareth321

That would be a hard no for me.

What a lovely idea. Maybe I should try this.

Cloudflare recently reported that the majority of internet traffic is bots now. It wouldn't surprise me if the vast majority of traffic and comments on Reddit is now bots. Those organic conversations you're having? Bots. Those arguments with annoying people? Bots. All those positive comments and stories about the latest product/movie/politician/party/company/country? Bots. The entire front page is most likely owned and controlled and paid for by the highest bidders.

I think people are becoming more cynical about "social" media. It's far less social now and mostly anti-social. Maybe the new term should be rage-media or consumption-media. Either way I don't need more outrage in my life. I say good riddance to social media.

I don't mind logging in, but when they remove old.reddit altogether, I'm just going to stop using Reddit. The new interface is the slimiest culmination of all the worst dark patterns and manipulative algorithms. I gave up on Facebook and I'll give up on Reddit. I have so many games and books and magazines to consume which I just don't because of this awful habit of picking up my phone to doom scroll. In fact, bring it on. Turn it off. I dare you. I need an excuse to log off Reddit forever. It's a toxic cesspit of the very worst people on the internet. Worse than 4chan. Worse than Kiwi Farms. When I look back on my time on Reddit I have almost only disdain. It is a supreme waste of my life.

Hitler was not elected Chancellor by the German public. Germany had/has a mixed-member proportional system, meaning that coalitions may form after everyone votes President Paul von Hindenburg appointed Hitler as Chancellor on 30 January 1933 after negotiations among conservative politicians who believed they could control him. The Nazi Party was the largest party, but it won only 33.1% of the vote in the final free Reichstag election of November 1932 and never secured a popular majority. Once appointed, Hitler exploited emergency powers and political intimidation to dismantle democratic institutions and convert his government into a dictatorship.

During his political ascent he used his Brownshirts - a large, organised paramilitary force - to assault political opponents, break up rival meetings, intimidate voters, and fight street battles, helping destabilise the Weimar Republic and suppress opposition.

I am also worried about this. Typical ads are already devastatingly effective against most people, causing incalculable distortions in market dynamics and incalculable losses to consumers. This is an order of magnitude worse.

But there is something even more insidious: elections. Ads will be purchased to shape public perceptions. Those with the deepest pockets will have ads which are charismatic, convincing, and unbelievably effective. This is the death of democracy if regulation isn't adopted right now.

Unfortunately this isn't straightforward. The distinction between a product, service, person, and party are impossible to clearly delineate. Is a SpaceX ad an endorsement of Elon Musk? Is an Elon Musk ad an endorsement of the U.S. Republicans? The famous Citizens United ruling determined that there is no practical legal distinction in this domain, meaning that without a Constitutional amendment (which will never happen in this political climate), the U.S. is going to barrel headlong into some kind of dystopia.

My only ray of sunshine here is competition. If OpenAI gets a reputation for untrustworthy AI, and there are viable competitors, people will hopefully switch. It's messy and imperfect and will require constant vigilance (because they're not going to tell us when they're misleading us). I think open weights models have become very good, and will improve much more over time. When we can reasonably run these models locally, ubiquitously, we have cost effective options.

Kago

Nice try, LLM! I'm onto you!

All of the scenarios involve circumstances we could not control. For example, someone pushing into our lane on the motorway in the blind spot, requiring evasive manoeuvre to avoid. It's possible Danish drivers are careless or aggressive, but I think it's more likely you just don't realise how many close calls and evasive manoeuvres you need to perform on an annual basis with average driving kilometers and regular defensive driving.

You cannot rely on the extra safety it provides. A driver with basic competence should be able to avoid most risks through anticipation before they happen.

It's too late. These safety features are saving lives all over the world, every day. As per the Danish and Dutch regulators, FSD is already safer than the average driver. I would prefer that all the drivers on the road are using this technology than manually controlling their cars. People make all kinds of mistakes which FSD does not.

OpenAI and Anthropic aren't even listed. Their failures would barely dent the stock market.

Bond sales are pretty normal for companies, AI or not. Leverage is in very low ratios compared to the dotcom era.

Much of the dotcom era telecom expansion depended on debt and projected future demand, but the resulting networks were so overbuilt that only about 2% of North American long-distance capacity was being used. Falling prices left numerous operators unable to service their debts. Today’s AI expansion is led mainly by already-profitable companies with large existing revenues and cash flows. Microsoft alone generated $136B in operating cash flow during its 2025 financial year while spending $65B on capex. It held $95B in cash, compared with total debt of about $43B. That means a disappointing return on AI investment would reduce profits, cash reserves, and shareholder value, rather than making these companies insolvent.

Even if OpenAI or Anthropic or both fail, they're not even listed yet. It barely makes a dent on the stock market. Their infrastructure and IP is purchased by Microsoft/Google/Meta, and those companies valuations adjust accordingly.

While I'm a lucky lifetime Plex pass holder, the service I get with my own self hosted setup is leagues ahead of anything I could pay for. I would probably pay for a Plex subscription if I started today. I hope Jellyfin improves its remote streaming UX, subtitle management, and app polish, because it's a pretty good contender otherwise.

IMHO I think the blast radius is implicitly contained. The major investments are in data centres, and the current investment cases impute near zero residual value after five years. Meaning that current valuations already assume "catastrophic" declines in equipment valuation. This is unusually clear-eyed and sober investment calculus in the tech space.

Further, unlike during the dotcom crisis, most of this spending is not driven by debt. It's mostly funded by the large companies which are producing enormous revenue and profit to pay for this. It's an order of magnitude different.

The major question mark on these valuations is the revenue assumptions, which can be reasonably criticised. A bear case here is that revenue growth is not as aggressive as projected, and valuations steadily decline over time. I don't see a likely scenario where the entire sector collapses. There is no apparent cascade failure mechanism. Of course, these mechanisms aren't always immediately clear prior to crashes.

We must remember that all of the models coming out of China are presumed to be distilled frontier models. Meaning a) they will always be x days/weeks/months behind the frontier models, b) they will never be quite as good, c) inference will generally be constrained by compute capacity (especially as the frontier studios have an incentive to capitalise on their moat), meaning Chinese studios will always be at a disadvantage.

The real wildcard here is self-improvements. It looks like we're already in the singularity, meaning a large proportion of LLM development is already done by LLMs. The development cycle on these might be months now, but it will be weeks soon. Days within a year, then hours, minutes, seconds, and milliseconds. It's impossible to predict what this curve looks like.

The UN and “international law” are not the final authority on anything. International law is a loose collection of treaties, trade agreements, conflicts, supply lines, and pinky promises. It only takes effect in a democratic nation if the people vote in a government which makes it a law.

The UN is a political organisation, not a neutral arbiter. During the Rwandan genocide, the genocidal government still held a seat on the Security Council while the UN reduced its peacekeeping force. Iran was elected to the Commission on the Status of Women. Libya and Russia were elected to the Human Rights Council. These appointments are driven by bloc voting, diplomatic deals and state interests.

International courts are not above politics or error. Their judges are selected through a system in which dictatorships and abusive states have a vote. The UN is not a democracy. While each member state might appear to have one vote, that does not indicate their power, influence, credibility, or moral standing.

Note how I did not make a judgement on whether Gaza is a genocide. I am merely explaining that deferring to the dictatorships and pinky promises is not a good argument.

It's not a secret that China is run as an authoritarian dictatorship. Xi Jinping cannot be ousted in an election. He appointed himself for life, and regularly purges anyone who might usurp his power. They don't have a functioning democracy. Are you seriously disputing any of this?

None of this excuses issues with American democracy.

Remember, these cars are running local LLMs, not frontier models. The issue with self-driving cars has been the edge cases. The 0.0001% of situations where the models did not have sufficient training data. This is compounded by the hardware limitations. Onboard RAM in a typical Tesla on the road is 16GB (+16GB for the backup computer). This has to run the existing onboard OS and other operations plus the LLM. These two factors combined means that the cars are currently incapable of negotiating the 0.01% cases, let alone the 0.0001% cases. And this is compounded by the fact that LLMs cannot currently update their weights in real-time, like humans. It takes months to train a new model. Special small models can be very tricky, especially around safety and mission critical applications like FSD.

All that said, current data shows that FSD is already better than human drivers on average. See the recent regulatory decisions by the Dutch and Danish road safety authorities. So we've already crossed the rubicon. All improvements now are icing on the cake. My prediction is that local LLMs will get much better, very fast. How that's operationalised with Tesla (or other) data is yet to be seen. They have at least three new ASCIs/SoCs in the roadmap for improved LLM efficiency and with a lot more RAM. Plus they just announced new technologies allowing the local LLMs to learn from driver intervention and behaviour. Some form of vectorised RAG, which could mitigate a lot of the limitations around real-time learning.

I am very optimistic for the future of self driving. I own a Tesla with FSD now, and it's incredible. It makes mistakes, but fewer than I do, and so far has saved my butt (and my wife's) several times from obstacles and emergencies we would not have seen. The car has undeniably made us safer.

Put all of these trends together, and I think that in 10-15 years, we are going to have consumer PCs (and phones!) running models doing pretty much anything that frontier models can do right now.

At current pace, we'll have open weight LLMs with frontier intelligence in 6-12 months. The constraint is RAM - both for the model and the context. It's likely that distillation and quantisation and TurboQuant will significantly reduce RAM requirements. I think we'll have Opus 4.8-like performance on 64GB of RAM in two years.

Of course, by then, frontier intelligence will be god-like.

It's strange because my impression has been pro-Chinese comments on HN. I suppose it depends on the topic, the time of day, and how many bots each government is employing on a given day for the key words and their compute budget. This submission seams rather balanced to me.

While Apple has technically permitted third party browser engines in the EU, they have made it so onerous and expensive that no company is likely to build one. Apple remains the gatekeeper and will likely deny any builds for many years while they abuse the EU courts in delay tactics.

The EU has been fumbling the application of the Digital Markets Act. They have been sitting on a judgement of non-compliance regarding third party stores for literally years now. By now they could have moved to forced compliance and huge fines. It's not clear if there is some kind of hidden diplomacy happening, a desire to not anger Trump, or sheer incompetence. Either way, I have lost hope that the EU bureaucratic engine will ever do anything to help ordinary people. Their only function now appears to hurt the very weakest in society, because they can't fight back.

-angry Dane

This is why I subscribe to Kagi. Google has been censoring topics for years, for many reasons. I have a few domains I use to test whether a search engine has been censored, and Kagi passes all of them. For example, many years ago, Google deleted the Proud Boys website from search. I believe they have deleted later iterations as well. No doubt there are millions of other website they've deleted, suppressed, or down-ranked. They're not transparent about this process or how the decisions are made. Of course I don't support or condone the Proud Boys. That is not the point. The point is that I should be allowed to search for information and make up my own mind.

I wish it were not necessary to use search indexes from multiple countries I do not like, but the American government and American companies have decided they should be in charge of what users are allowed to see. I see no alternative.

When I look a decade ahead, I think the advertiser internet model is going to die. People won't be using search engines anymore. They'll be using LLMs. They won't be reading blogs anymore. They'll be asking their LLMs. They won't be looking at review or recipe or travel sites anymore. They'll ask their LLMs.

Of course this creates second order effects. If no one ever views the ads on the review and recipe sites, those sites will disappear. Then what do the LLMs reference? Parametric knowledge? Google is trying to get ahead of this with "Google Zero:" an internet in which one never leaves Google because they cache and serve everything a user could ever want. All stolen from other websites, of course.

I hope we end up back at the internet of the 90s. Small, unmonitised hobby spaces. Unfortunately bots will be 99% of all internet traffic within a few years, so how do we protect these spaces from unsolicited spam, disinformation campaigns, trolling, etc? We might actually go and touch some grass and re-learn how to socialise in person. This is not a bad thing.

Put it in a faraday cage with some kind of explosive device set to trip if it escapes containment. It's the only way to be sure.

Musk had around $180M when he founded SpaceX in 2002, and purchased Tesla in 2004. ChatGPT tells me that more than 55,000 people in the world have more wealth than this. So either people who have money choose not to make such risky investments, or they're incapable of it. Probably both. We're very lucky that someone who is both capable of building companies like this and is willing to go bankrupt doing it exists.

This is about gender, not sex, and that's not the definition I used for "gender" that you agreed with. You agreed with this definition of "gender": "female gender refers to the social roles and norms associated with being a woman".

According to the definition you used, if a woman likes doing masculine things, does that automatically make her a man? I presume not. There is another part of the definition which you omitted.

Your brain creates your perception of your gender. But trans people don't have a "female brain". There isn't a single aspect of your brain that determines your gender; there isn't really a commonly accepted scientific definition of a "male brain" or a "female brain."

So what is the thing which determines gender then? What is it, exactly, beyond doing masculine and feminine things? Surely our identity is distinct beyond the things we do?

I'm not American. I think Donald Trump is a clown. I'm so tired of your political team sports. If I criticise one of your sacred cows that means I'm on the enemy team. There is zero room for critical thought or nuance with you.

This is interesting, thanks!

You're missing the point. "Disagreement" implies some level of care. You can't claim to not give a fuck, and also claim to disagree. These two statements are incompatible.

In this context when I write don't "care" I mean I respect your right to live how you like, provided it doesn't encroach on my right to live how I like. I obviously have opinions.

Premise (definition): To have the female gender is to follow the social roles and norms associated with being a woman — so A implies B.

No, A does not imply B. Following social norms common to one sex doesn't imply one is that sex. This is a fairly milquetoast logical proof and I'm surprised to see you making such a basic mistake. The feminist definition is a role, not an identity. The core of feminist theory is that women are not the sum of their social role. The exact opposite of the theory around transexualism.

The left didn't make this a wedge issue. This is obvious; there is nothing for the left to gain from doing that. The left did the same dumb thing they always do: take a moral stance and then get kicked in the teeth by bad-faith polemicists who'd rather talk about bullshit that never happens to anyone but sounds kinda scary than about actual shit that actually happens and makes them look bad.

For me the benefit appears obvious (and cynical): power. For years, people could be summarily and permanently banned from social platforms, banks, be fired, rejected from official roles, etc, because they stepped out of line on this issue. It was a very clever power grab by the unscrupulous who used it to attain power in organisations and society by hijacking people's empathy. If I'm being fair, there is also an ideological component, meaning some people genuinely believe that males can be born with a female brain/soul, and vice versa.

We’re talking about the definition of words and how censorship is a tool of authoritarianism.

No one can describe a single moment a society becomes authoritarian, and asking such a question is obviously specious. The devolution happens incrementally over long periods of time.

Yes, only to then go on to write stuff like "disagreeing with the way you live". I don't know if you're lying to yourself or to me, but you very obviously, very clearly do give a fuck.

That’s, like, your opinion, man. One can disagree with you and still believe you have the right to live how you like. At the same time.

Right, so you agree that the female gender refers to the social roles and norms associated with being a woman. So anyone who follows these social roles and norms has the female gender.

No. Your first sentence doesn’t imply the second. A society can have gendered roles whereby men like mowing the lawn and women like cooking, but it doesn’t mean that men who like cooking are women. Feminist authors use gender roles to describe oppressive patriarchal expectations placed on women.

Honestly, this is a fucking stupid discussion. In the end, it doesn't matter. If somebody asks you to treat them a certain way that costs you absolutely nothing and makes their lives better and you don't, you're just being an asshole. In all likelihood, you will interact with maybe a dozen trans people in your whole life. If you want to be an asshole to people in those dozen times, that's your problem, but it really doesn't matter to your life either way. All you achieved is being an asshole.

As long as we agree that being an asshole shouldn’t be against the law then live your life. We can each agree the other one is being an asshole and get on with the rest of our lives.

For the record, this wasn’t an issue for me until the left made it a wedge issue. I’ve had a trans friend since well before this became a political football. When the left attacked for disagreeing about the treatment, especially with regards to children, is when it became an issue for me.

If the model could substantiate that with facts, fine, but it could not.