This is a meme and it is unfounded. In most cases, nature is good enough.

eg: https://news.ycombinator.com/item?id=41024612 (9 days ago)

Since then they're mostly rapidly auto-flagged.

i know that this is probably a very bad approach, but until now it worked out - if a text contains grammatical errors i automatically assume that it was written by a human, as AI-content always sounds so polished, boilerplate-y and corporate. but you can probably tell your AI to integrate errors, ignore capitalization and miss out on some punctuations.

I asked ChatGPT to reply to your comment and to include some errors, leave out some punctuation etc., that's the result:

oh, i see what you're saying. yeah, it's really hard to tell if something is bot-written these days. like, even reading your comment, i start to question if i'm talking to a human or not. the examples you mentioned do stand out, but more because they're so similar, not necessarily bot-like. and you're right, the "firehose of misinformation" is just getting worse and worse. it's a strange new world we're living in.

i think we're cooked.

But what is it about biodiversity that you like? Your comment seemed to imply you believe it has some provable value or utility that you were unable to articulate (which I'm not saying it doesn't). But if value or utility isn't why you appreciate biodiversity, then why do you feel a need to justify your position?

I "cheat out" writing my comments. They're mostly for the audience, and only a little bit for the person I'm replying to. Do you write for the person you're replying to?

"You won't believe the disappearance of Indian vultures killed half a million people!".

Better?

But we have almost lost Florida as an orange producer due to the fragility of a monoculture against disease. So in some ways it is even worse. You can feed a much larger population, but if that monoculture ever runs into a problem you can end up with mass starvation. See also: the Irish potato blight.

Even those monocultures depend on a working ecosystem around them.

Regarding yields, it’s a risk assessment. They can be great in the short term and then crater when the soil is destroyed. At this point fertilisers are required just to keep production level. And if there is a disease that wipes out a species, then it’s game over. And it happens occasionally, from the Irish potato blight, the almost-complete destruction of European vines, whatever is destroying olive trees near the Mediterranean. There are several examples. Lack of flexibility in the long term means lack of resilience.

> Resilience is another thing that's very hard to reason about, because why would resilience matter to you if your race dies out? Sure, some animals and insects would have a higher chance of survival under different settings, but why does that matter to you, a human?

There are philosophical problems with this (those species are not less deserving than we are), but let’s put them aside for the sake of the argument.

The problem is that there is a lot that we don’t understand about the world around us, and we occasionally discover that a species was useful when it disappears. Or the contrary, that it is an invasive pest if we introduce it somewhere. Or that useless things like mangroves are actually critical to avoid unchecked erosion. Or that burning that useless Amazonian forest is actually terrible on at least 3 levels (direct emissions, that forest is not available anymore to absorb other emissions, topsoil erosion and degradation that makes it terrible agricultural land over a generation).

This is very bad because we have only one planet and we cannot shrug, write it down, and do it better next time.

Resilience of your food supply should matter to you, as an organism that needs food to live.

These monocultures and the carbon-based energy systems and the capitalist systems that depend on will fail and millions will die.

We are heading for a cliff.

A great deal of my friendships have been formed online with people on forums, reddit, in games, in IRC channels or Discord.

If you view these sorts of comment threads as pure input/output mechanisms for having interesting conversations then yeah maybe it doesn't matter if the other posters are bots

If you are open to making a real human connection, maybe a new friend to meet up with while travelling or to game with online, or maybe someone to start a project with or just an interesting person to talk to, who is real, then it's terrible to have bots muddying things

Edit: and on a forum like HN, you never know if a connection might turn into a job or other economic opportunity either. But they wont from bots

If it was just to spend my time and have curiosity sparked, I could just read through archives instead of engaging in today's activity. But I don't do that because I'm looking for immediacy. Human, social, original, immediacy.

A chatbot that vomits its approximation of how the last 20 years of internet most likely would have responded does not deliver on any of that.

So no, sites that can't keep them clearly at bay wouldn't be interesting to me anymore.

People love reddit to get genuine feedback on a product. Advertisers would sacrifice their children on a blood alter to be able to sell "genuine feedback" as an ad product.

As far as bots providing useful comments, xkcd did it (https://xkcd.com/810/) in 2010, but there are downsides. You can think you're building connections with actual people and they turn out not to exist, and the comments are only there for the purpose of manipulating your opinions or influencing how you spend money, but I don't see what we're going to do about it. Captchas certainly aren't a viable option anymore.

ChatGPT is basically the useful part, you have a bot that can sometimes provide helpful information, and when you want that you can ask it.

Yes, because they used to manufacture consent. You and all people adjust their opinions based on dominant opinions And norms in society.

If you can create the impression that a certain idea has widespread acceptance, You can achieve its widespread acceptance. Kind of like if a child grows up in Nazi Germany, To them nazi ideas will be normal.

There's never been a time in US history or a place in the US such that there was a shortage of food for people who had money to buy food unless you count situations like the Donner Party in which a caravan spent the winter of 1846–1847 snowbound in the Sierra Nevada mountain range.

The US not only has the most productive chunk of farmland in the world, but also much fewer natural barriers to efficient transportation compared to productive farming regions in the rest of the world. Certainly during the Dust Bowl, there was plenty of food grown in places like Iowa and Illinois that could easily and reliably have been shipped to Oklahoma, but no one did because no one (not even the federal government) considered it their job to help the hungry people in Oklahoma.

Part of the reason it took so long for the US to grow a welfare system is the American ethic of individual freedom and distrust of government, but another part is that it was easier for people to get the basics of survival than in places like Germany where the welfare system developed many decades earlier (the Dust Bowl being of course an exception to the general easiness).

chatgpt-4o after much hype, billions of dollars and years later didn't improve much in terms of core intelligence of its responses.

I guess thats not enough for "AGI is coming because we have chatbot" crowd.

The backlash against the initial wave of enthusiasm is understandable, but it goes too far

I think it’s a mistake to equate the language ability of these models with general intelligence, even if the language produced is excellent.

I think some people see AGI that way (some nebulous unfalsifiable thing), but that’s not what I’m arguing. I think there’s a strong case to be made that at a minimum, AGI’s core “knowledge” will need to be made up of far more than just a lossy textual representation of the world.

Humans come in all levels of intelligence. Yes there might be a human somewhere that believes that they should eat rocks or add glue to pizza. But as a generality we can assume that adult "humans" know not to eat rocks.

But don’t take my word for it. Go spend some time experimenting with the state of the art and run some of your own tests with it. Assuming you yourself are not one of these bots, what you find might surprise you.

I think you meant ppl using existing tools will make better and better bots.

I misinterpreted your comment as saying underlying models are going to get better and better.