It is good I saw photos of platypuses before Stable Diffusion.
A.I. imagery may destroy history as we know it
131–140 of 159 posts
Re: A.I. imagery may destroy history as we know it
#132Earlier quoted context omitted.
We are discussing whether the article has content or not, clickbait is a particular type of low information content on the internet which I think is an incorrect characterization of this article however I do agree it's low information, and basically the premise of the article was reverse engineered by a writer that thought "wow these photos are cool." If you find my comments on the matter also low information, I can…
I think, it's about the kind of fascination with an image that connects us to a greater narrative that is our identity. The notion that these generated images do fascinate, while connecting us to a non-existing narrative, is neither out of context nor just superficial, nor is so the exhibition of these images. The fascination emanating from these images (thanks to the transfer of textures and hints of authenticity) i…
Provenance, patina, narrative all of these live inside your head and have nothing to do with reality. It’s the perverse motive of the collector seeking unchanging authenticity. Everything and everyone evolves and changes over time. You can’t stop the clocks.
Re: A.I. imagery may destroy history as we know it
#133Maybe a crypto person would know more, but couldn't some DAO be incentivized to verify the authenticity of current known historical documents/images and create a block chain of agreed upon authentic historical imagery? The longer it takes for us to do that, maybe the more likely falsifications can slip in, but I have a feeling that the real world nature of many history books and libraries means we can have a pretty s…
you still have the issue of "junk in, junk out". Being on a block chain does absolutely nothing to ensure that the information on the chain is correct. There isn't really any reason to use a DAO for this vs a standard database.
Re: A.I. imagery may destroy history as we know it
#134Earlier quoted context omitted.
Your history classes sound a lot more useful than the ones I sat through in the US - ours was mostly delivered as a set of settled fact.
There is a local school that is dedicating 90 minutes to math and 90 minutes to english. Splitting science and history to a mere 45 minutes each. I completely support this. Science prior to college was basically a waste. History was full of bias and axes to grind. Don't get me wrong, teach these subjects, the scientific method is great. However, the value is low compared to the tools given in Math and English.
I don't know how you could expect anyone to really understand literature without history. Imagine trying to explain Animal Farm without Stalin.
Re: A.I. imagery may destroy history as we know it
#135Earlier quoted context omitted.
Your history classes sound a lot more useful than the ones I sat through in the US - ours was mostly delivered as a set of settled fact.
There is a local school that is dedicating 90 minutes to math and 90 minutes to english. Splitting science and history to a mere 45 minutes each. I completely support this. Science prior to college was basically a waste. History was full of bias and axes to grind. Don't get me wrong, teach these subjects, the scientific method is great. However, the value is low compared to the tools given in Math and English.
> Science prior to college was basically a waste.
My public school biology class involved sequencing and editing DNA to make bacteria change color.
I would not call it a waste.
In fact I wish I would've had more science in school, I managed to go until college until I took my first Physics class, and physics is really useful.
Re: A.I. imagery may destroy history as we know it
#136Earlier quoted context omitted.
There is a local school that is dedicating 90 minutes to math and 90 minutes to english. Splitting science and history to a mere 45 minutes each. I completely support this. Science prior to college was basically a waste. History was full of bias and axes to grind. Don't get me wrong, teach these subjects, the scientific method is great. However, the value is low compared to the tools given in Math and English.
What tools are you given in English? And how many of the tools in Maths are you still using? Like, I’m very happy that I got to learn probability theory, but all the differential equations that broke my mind seem entirely pointless now.
This might feel less useful for CS, but for other sciences and even engineering (mechanical, aero, materials, chem, electrical, marine) you need differential eq.
And than natural sciences in general (climate, space) also have applications there.
Subjects that feel useful post graduation, if they are being taught in high school early college the time gap makes it less useful. HS and early college classes fundamentals are needed for the specialized curriculum that needed for higher college and post graduate education.
Re: A.I. imagery may destroy history as we know it
#137Earlier quoted context omitted.
> Why would any kind of camera photo signing algorithm fare better? You probably wouldn't even have to bother. If you want to fake history, just launder your images: setup rig where an SLR camera can telecine the images off a high resolution screen. With some optics you could probably also adjust the other camera parameters (like focus, f-stop) to make it more convincing.
You could imagine a timestamp being built into the signature, and maybe even gps coordinates, although I agree it wouldn't be a complete solution.
> You could imagine a timestamp being built into the signature, and maybe even gps coordinates, although I agree it wouldn't be a complete solution.
Both those things can be faked. IIRC, they make chambers right now that simulate GPS signals for testing receivers. If you can fake GPS, you can fake the time, too.
Re: A.I. imagery may destroy history as we know it
#138I mean outside of the two World Wars, practically all the history I was taught in high school was pre photography.
Surely you’re not going to be fooled by a photograph of Julius Caesar.
Re: A.I. imagery may destroy history as we know it
#139Meh - I don't know. At least in the UK, "History" classes at school were less about memorising the facts, and more about questioning the who, what, why of some source of information. One thing that stands out in my memory is multiple different teachers always drilling into us: "never trust photos" (you don't always know who took them, for what reason, and if they were real or staged or whatever) If 12 year old kids a…
Yeah, the ”post-truth era” is a weird catchphrase which has been going strong for the last 5 years or so at minimum, yet the described dystopia has never arrived and never has been on track to arrive.
Re: A.I. imagery may destroy history as we know it
#140Earlier quoted context omitted.
eyeroll I read the economist on a weekly basis and can confidently say that that form of curated professional journalism is much better than random partisans on twitter loudly shouting about everything
This is somewhat hilarious because the Economist is famous for its unsigned articles and IIRC lacks citations, such that you have no way to verify whether the person writing them has any expertise in whatever they’re spouting off about. It’s effectively a middle-brow subreddit in print.
While I agree that the Economist is infamous for having the occasional-to-frequent terrible take under a common pseudonym, I think you're wildly overselling the quality of most mid-brow subreddits (Or have a definition of "middle brow" that only includes the top 90th percentile).