261–270 of 452 posts
Honestly this feels like an old man screaming at the wind type thing. Inevitable doesn't even begin to describe was going to happen over the next 20 years. You should be much more terrified of when their content isn't garbage.
The answer is not to "disconnect". The answer is to read critically, to read for argument, and to try to connect with the intent of what the author was saying. All the while realizing that most of the content out there is bullshit( https://en.wikipedia.org/wiki/On_Bullshit ), whether written by a human or not. And yes, the amount of bullshit is about to increase exponentially. But it was already a majority. Ironicall…
Our attention, mind space, and time are all limited. We just don't have time to process this exponentially growing pile of content.
Earlier quoted context omitted.
> Another problem is AI training on AI generated content. This could lead to degradation of results. That's only the first step in the degradation. In a worse case scenario, where AI-generated crap is not contained, imagine this: young and still immature humans "learning" their styles of writing, information, and social expectations from content generated by AI that itself trained on AI-generated content─and then mor…
Luckily, the phenomenon of puberty will make sure that young people absolutely do _not_ want to comply ;).
We should ask questions like "Can countless AI's fracture society into a million sub societies disconnected from the reality around them".
Earlier quoted context omitted.
First, Monsanto and its way of operating very much still exist. Just because it was purchased by another company (which of course resulted in an even greater monopoly), does not make its impact to the world disappear. Second, everytime this topic comes up, pro-GMOist make it sound like the technology is making some great strides with providing 'healthier food for everyone' with no downside. Why is it that the current…
> Why is it that the current model for GMO aims to increase resistance to pesticides instead of simply growing the yield? I It doesn't; there are plenty of traits beyond pesticide resistance that have GM crops targeting them, including yield, drought resistance, nitrogen fixation, etc. Pesticide resistance is the most established on the market becauae there happened to be particular success with it decades ago , that…
This might be the case, but still my original bone to pick remains. GMO is not without downside, and people cheering for it without a hint of scepticism reeks ignorance and unwillingness to learn from past mistakes that were made in the name of science.
As you've highlighted, said technique was used for decades, for the detriment of soil health, bees and all kinds of smaller organisms, without it being questioned.
As for your last paragraph, I'm not sure I understand. Are you saying that voicing these 'uneducated' opinions about GMO simply pushes the crops to incognito mode, making them less visible for consumers? Better to see your killer face-to-face than being stabbed to the back? Not sure about that.
What I really wonder is: is there a correlation between consumption of streaming tv series and positive attitude towards this "AI" wave? "statistical mean of everything uploaded by humans on the internet which fits certain criteria" applies to both ChatGPT/StableDiffusion and most movie series on streaming TV. My pet theory is that if you can stomach watching the same stories with different skins hundreds of times, y…
I’m 41 and I’m utterly baffled by my 24 year-old SO’s interest in binging an apparently infinite litany of TV shows from streaming services. I’ve watched the first six seasons of Game Of Thrones, the first season of Dark, all of The Expanse (reading the last book of the final trilogy now), His Dark Materials (because I read the books) and Andor. That’s it. I watched some Star Trek TNG & DS9 as a kid and was horrified…
I see parallels here with industrialization and automation in manufacturing and food production. Today, it's self-evident to us as consumers when a product is cheap, low-quality, manufactured garbage from China. In many cases we can choose to spend more and buy a handmade or custom version of the product we want. In the future we might knowingly consume AI-generated cheap text for some purposes and opt for "the real…
> Today, it's self-evident to us as consumers when a product is cheap, low-quality, manufactured garbage from China.
Correct.
> In many cases we can choose to spend more and buy a handmade or custom version of the product we want.
Not quite. It's increasingly hard to find a version of a product that's not low-quality garbage - the expensive option is often just the cheap option with extra markup for those who got fooled by marketing, or naively believe companies still care about their brands.
That's the unexpected side effect that, I feel, people still mostly failed to realize: because running a company and manufacturing anything has nontrivial costs, letting the market race to the bottom hollows out the middle. You're left with a bunch of specialty vendors who make quality goods for other businesses or governments, at ridiculous prices, and the low-quality throwaway garbage for everyone else. The "more expensive, reasonable quality" product group disappears - the vendors move either into low-quality crap or too-expensive-for-individuals specialty items (or sometimes both), as the middle can't compete with either end.
I imagine the threat of AI-generated content to be similar: as it completely blows the bottom across creative industry, it will hollow it out. Skilled creators will, when they can, move to the high-end Big Art, where you pay more for the work's provenance than for the enjoyment itself, because they won't be able to afford competing with somewhat worse but much cheaper AI-generated shovel-art.
(Also relevant is that regular people are usually as good at distinguishing between crap and good products as they're at distinguishing between crap and good art.)
Earlier quoted context omitted.
GMOs are terrible which is why they're banned in most European countries. > bad wrap they've gotten from science ignorants Science is not about blind trust. How many times have scientists discovered wondrous new things like asbestos, Thalidomide, leaded gasoline, teflon, only to take decades to realize the consequences? Even if we blindly trust science, there are various other issues like: patenting life, monopolizat…
> GMOs are terrible In what way? > various other issues like: patenting life, monopolization, reducing genetic diversity, GMO cross-pollination Agree these are all terrible, but nothing there is GMO-specific. That's all either bad agricultural practices like mono-cropping that can be done with any crop, or societal problems related to intellectual property or markets.
> In what way?
Because they're the exact equivalent of closed source in software.
There would be means to research, produce and use GMOs in good ways, but the way our economics work, they can't be used without either harming natural ones and/or involving NDAs, lawyers and all sort of things that in some contexts make closed source a nightmare for pretty much everyone except those selling it.
It is a real problem. The web will soon be full of AI generated bullshit. What might come back as useful tool? Authorship. People actually signing the content and putting their reputation on the line. It still exists obviously (journalism…) but it might need to be standardized and industrialized.
"Fall, or dodge in heaven" by Neal Stephenson actually touches on this topic quite a bit and comes to a similar conclusion. Not his most accessible work but I enjoyed it. That book was written just before Trump won 2016 elections and in hind sight making that book really spot on and timely.
In the book, different streams of truth become so hard to separate due to all the spam, automated content, etc. that people start filtering it based on reputation and signatures. The game then becomes curating and filtering the content, which is ultimately not an objective thing but a subjective thing. So the net result is what we are actually starting to see today which is different groups of people locked in their own bubbles and no longer even getting exposed to the other side's point of view or even able to agree on some basic facts. Like who actually won the election. And people are feeding those bubbles with nonsense to influence opinions, get people to vote certain ways, or simply to just create chaos and fear.
In the book the topic is not an election but the seemingly simple question of whether or not a certain town in the us still exists after it was supposedly blown up with a nuclear bomb. In the book, this event leads to people acting as if that happened and ignoring all evidence to the contrary pointing out that in fact this was an elaborate hoax designed to create a massive panic.
That particular dystopian outcome is not necessarily what would happen. But using technology to authenticate where content actually comes from along with some reputation mechanism would go a long way separating the noise from the genuine content. It's simple and easy to do. So, why don't we?
In just the past couple of weeks, I've noticed lots of machine generated cooking recipes. They follow the typical SEO pandering blog format, where there's an unnecessary story about the author's trip somewhere or their child's school play being the inspiration for making this particular recipe, followed by the actual recipe details. The machine generated stories are even more pointless and meandering than what humans…
Whenever this technology is used in an accuracy-critical setting, it is a recipe for disaster.
These "AI systems" produce the MOST PREDICTED output whenever there is any ambiguity, and this REPLACES the ACTUAL UNEXPECTED value/ITEM. In information system terms, they are literally stripping out content with the highest information value.
I saw this first hand giving a deposition that was supposedly transcribed by an "Artificial Intelligence" system. The output looked beautiful, until I started checking it for errors. Normally an errata sheet has a dozen lines for errors. Mine had pages. And they were substantial and very meaningful. The most obvious site was a specific name that sounded vaguely similar (started with a "J" and "K" sounds) was replaced by "Jack Kennedy". It also replaced "you have" with "I have" and omitted "not" several times, literally reversing the meaning of the statement. There were 85+ errors that I caught, over an order of magnitude more than normal.
These things are basically fine as glorified type-ahead systems where it doesn't matter much and a human is in the loop. But using them in any situation where the results are critical is, at this point, basically fraud. They do not do what the proponents claim.