No amount of big context window can stop the model from context poisoning. So in a sense, it's a gimmick when you start having the feel of how bad the output is.
I'd imagine India has some pretty insane stuff driving around so that's not surprising. The US effectively did that (arguably even more extremely) 15 years ago with the "cash for clunkers" thing.
There's this bizarre alternative history around "cash for clunkers" where an incentive program has been rewritten as this great evil that forced older vehicles off the road. All participation was voluntary. While internet commentators bemoan the "classic" cars scrapped in the program, these were almost certainly in terrible condition - people with classic cars in good condition know their value (or at least the party taking it as a trade-in does).
The author definitely is not being very knowledgeable. In the comments, they didn't try Zen because they assumed, it can't sync bookmarks and extensions, which in fact it can and has Mozilla account baked into it.
fwiw though: Zen does have other challenges at the moment with the Widevine licence. so you effectively can't use it to watch most video services today.
But point taken, from a technical accuracy perspective.
You are incorrect in your assumption. Though I would also like you to search for "IRC books reddit". Unlike Anna's Archive, you get high quality books with fast download speeds.
They don't just censor, they limit organic influences. Your content won't get displayed more than n times, so you can't get more popular than n views, unless the system selects you as today's lottery winner, in which case it will be (reported as)viewed trillion times.
The only defense against this is the fact that Twitter users know system too well for this to be not immediately obvious.
