chimerror

I'm Kitty (and so can you!)

  • she/her

Just a leopard from Seattle who sometimes makes games when she remembers to.


mcc
@mcc

Ronald Reagan came into office promising he would defeat the Soviet Union. He believed the way to defeat them was military might. He would ramp up spending on nuclear weapons and technologies like ABMs ("Star Wars") which, logically, would in the long run have no effect except to force both sides of the nuclear detente to massively increase their nuclear stockpiles to overcome the effects of the other side's ABMs and larger stockpiles. Reagan would increase America's military involvement in proxy wars against communism, and its overall military spending.

None of these missiles were used and none of this appears to have had any real-world effect. But there's a narrative you hear sometimes where Reagan's promise came about as a result of his actions, but purely indirectly and accidentally. The Soviet Union did, in fact, collapse three years after Regan left office. In the narrative, all of Reagan's military acts were useless or counterproductive but the Soviet Union's response was actually fatal, the Soviet Union freaked out in response to the military buildup's existence and ramped up its own military spending catastrophically— something the United States could at that moment afford by borrowing against its future but the Soviet Union, facing multiple internal crises that were at that moment coincidentally coming to a head, could not. In the narrative, Reagan achieved nothing but dumb-lucked the Soviet Union into destroying itself.

Is this narrative true? Never mind that for a moment. Just keep the narrative in mind.

There is a lot of noise in the press right now (I will not bother to link any of the "open letters") about "AI". We are promised (oddly enough, by the people making the "AI" software?) that AI is an existential threat to humanity that will one day destroy the humans. AI, goes the promise, will grow in power and intelligence until some allegedly logical chain of inescapable inferences force it to act out every science fiction dystopia simultaneously.

But what do we see in the real world?

The Internet is being rocked this week because Reddit's users are revolting and taking down their content, which has left Google mostly useless. Reddit's users are doing this because Reddit's owners are taking away the moderators' third party tools. Reddit's owners are doing this, if you carefully read their duplicitous statements, because the owners have convinced themselves that the third party tools are harvesting data for "AI" and the owners want a cut of the "AI" money pie. Reddit doesn't currently use any "AI". But because "AI" exists, they are destroying themselves.

Outside Reddit, indirect consequences of "AI" abound. Google's search results (outside Reddit) are becoming less and less useful as the operators of well-SEOed sites mistake autocomplete bots for reliable sources of content. Bing is trying to cut out the middleman and have their search engine present the results of the autocomplete bot directly, with no search results needed. Google— in a way that reminds me of how, once, their freakout about needing to compete with an ascendant Facebook lead them to rushjob "Google+" integration everywhere in a way that temporarily or permanently broke otherwise-useful tools like YouTube or Reader— claims they will soon be following suit with "Bard". In fact every single piece of software you might want to use seems to now be shoehorning dubiously-useful "AI" capabilities, all of which shuttle information back to a single dubious company named "OpenAI", and the software houses seem to be of a mindset that if users don't opt in to the OpenAI, that just means they need to try harder to force the feature. I've already had to stop using my favorite Android keyboard because Microsoft wedged in AI chatbot integration. In a keyboard. No sooner had I stopped using the keyboard than Microsoft announced they would soon be wedging similar integration into their operating system, which I can't stop using so easily. Soon you may have no choice but to use these "AI" tools which, I would like to stress, do not seem to work.

Reagan's nukes were bombs that never went off. Every nuclear weapon built in the US after August 9, 1945 could have been an inoperative blank with a hunk of lead for a core and the weapons could have had the same effect. We could live in a universe where nuclear weapons never worked at all, America just firebombed two cities in 1945 and did a theater show to fake the world into thinking they'd built a "new-type bomb", and that universe has a possible narrative where 51 years later the fake weapons tricked the Soviet Union into destroying itself anyway.

Do you see where I am going with this?


You must log in to comment.

in reply to @mcc's post:

imo, the AI stuff is (for now) a symptom rather than a cause

the real core cause is that with the Fed rate rises and such, the investment environment is getting harsher and venture capital is drying up, for reasons i don't care enough to actually understand

since most of the tech ecosystem runs at a loss and depends on VC to make up the difference, and even the few profitable companies are plowing massive amounts into risky bets as their core advertising business declines, this is big Panic Time for the entire industry

but at any given time, there's one or two magic buzzwords that'll loosen investors' pockets and get the VC money flowing. because of its high dependence on VC, the industry has always placed a disproportionate focus on those buzzwords even if they were obviously useless bullshit (for example, NFTs). but in the current crisis, announcing an AI product and making it as visible as possible is incredibly valuable. and given that the AI craze might actually leave behind a few useful products and applications after investors move onto the next big buzzword, it's at least less of a useless money furnace than the crypto crazes were

As for the Reddit stuff, they're just lying when they mention AI. The other big impact of the VC crunch is that sites that have happily run at a loss for years, like Reddit and Discord, are starting to feel like they can't reliably subsidize their losses with investor cash anymore. So they're getting ready to start aggressively monetizing their userbases. And for Reddit, which largely monetizes via advertising, part of the groundwork for this is forcing users off of third-party interfaces that Reddit doesn't have full control over. APIs don't serve ads, after all.

"The Zap Gun" future—weapons systems have become increasingly flashy and superficially impressive and, secretly, fake; the ultimate purpose of the weapons industry in Phil Dick's future is to set trends in consumer goods, and the weapons don't necessarily have to work at all.

to connect this up with the "AI" craze...surely one of the greatest attractions of the OpenAI business is that undoubtedly heaps of contracts just waiting to be signed with law-enforcement and military agencies. it's just the sort of magical-thinking technology they're bound to lap up, especially the cops, who would love to be able to pretend that "AI" can predict who's safe to throw in jail ahead of time. ~Chara