I think the pronunciation, specifically the blending of the end of “upside” and beginning of “down”, turns it into one of those compound words that your brain interprets as an independent word, rather than a combination of its composite parts.
You haven’t really looked into multi-agent setups at all, have you? Basically any system of multiple agents can double-check themselves.
Additionally, none of this conflicts with my original point. If you train a human on bad data, they’ll GIGO too. I know plenty of humans who have confidently told me objectively false things because they had bad training data.
LLMs are AI. They are not AGI. AGI is a particular subset of AI, that does not preclude non-general AI from being AI.
People keep talking about how it just regurgitates information, and says incorrect things sometimes, and hallucinates or misinterprets things, as if humans do not also do those things. Most people just regurgitate information they found online, true or false. People frequently hallucinate things they think are true and stubbornly refuse to change when called out. Many people cannot understand when and why they’re wrong.
Move fast and break things, I guess. My take away is that the genie isn’t going back in the bottle. Hopefully failing fast and loud gets us through the growing pains quickly, but on an individual level we’d best be vigilant and adapt to the landscape.
Frankly I’d rather these big obvious failures to insidious little hidden ones the conservative path makes. At least now we know to be skeptical. No development path is perfect, if it were more conservative we might get used to taking results at face value, leaving us more vulnerable to that inevitable failure.
That assumes that the population of these counties is significant compared to the cities though, right? These seem to be the lowest population-density counties in the state.
First off, this is obviously a sticky topic. Every conversation is controversial and speculative.
Second, I don’t really see a lot of legitimacy to the “gateway” concept. The vast majority of people use some variety of drug (caffeine, alcohol, nicotine), and that doesn’t really reliably predict “harder” drug use. Lots of people use marijuana and that doesn’t reliably predict hard drug use. Obviously, the people who use heroin and meth have probably used cocaine and ketamine, and weed before that, and alcohol/caffeine/nicotine before that, but that’s not really a “gateway” pipeline so much as paying through finer and finer filters. As far as I know, the concept has fallen pretty heavily out of favor with serious researchers.
In light of that perspective, I think you have to consider the goal. Is your goal to punish people, or to reduce the number and severity of victims? Mine is the latter. Personally, I think this sort of thing peels off many more low-level offenders to low-effort outlets than it emboldens to higher-severity outlets. I think this is ultimately a mental-health problem, and zero-tolerance mandatory reporting (while well-meaning) does more harm than good.
I’d rather that those with these kinds of mental issues have 1. the tools to take the edge off in victimless ways 2. safe spaces to discuss these inclinations without fear of incarceration. I think blockading those avenues yields a net increase the number and severity of victims.
This seems like a net benefit, reducing the overall number and severity of actual victims.
Those are distinct points from the one I made, which was about the characteristics listed. Sentience and sapience do not preclude a propensity to
regularly ignore directions, restrictions, hallucinate fake information, and spread misinformation because of unreliable training data
Llms are fucking stupid. They regularly ignore directions, restrictions, hallucinate fake information, and spread misinformation because of unreliable training data (like hoovering down everything on the internet en masse).
I mean, how is that meaningfully different from average human intelligence?
Her is set in 2025
Reflecting on your mistakes lets you learn from them and not repeat them. Reducing the number of mistakes you make is good for survival. Sorry, this is a feature not a bug.
“It would be nice to develop an auxiliary sign language to bridge the accessibility gap between the hard of hearing and those who don’t learn a dedicated sign”
“You’re just as bad as the colonizers that decimated native American cultures”
Get out of here with that bad faith savior complex nonsense. Teaching indigenous people English wasn’t the problem, the problem was beating children for using their native language. I guess you think literacy is racist too because literacy requirements were used to disenfranchise black Americans, huh?
Your sanctimonious colonization comments are dripping with irony. I asked a question, directly to another person, about their opinion of the concept as a deaf/hard of hearing person. You interceded uninvited, deliberately ignored the explicitly stated context of the question (gestural languages having unique properties from verbal ones) so you could shoehorn in your opinion about a topic explicitly excluded by that context, which you smugly assumed I wasn’t familiar with, purporting the relevance by referencing authors who wrote very little about the actual topic at hand.
You want to talk about colonizers, look at your own actions here.
My goalposts are in precisely the place they started: a collection of basic international gestures to facilitate the most basic communication. Where are you jumping to colonization? Where did I say that my cultural group gets to decide what the signs are? You’re, again, wildly overestimating the scope of my proposal and jumping to ridiculous, unsubstantiated conclusions.
You get a group of signers from around the world to develop an international pidgin (like they already do informally at international gatherings) and come to consensus based on commonality. When the majority agree on a sign, use it. Where there’s little agreement, choose a new sign. No finger spelling, no complex abstract concepts, just a formalization of gestures most people could probably figure out anyway. I fail to see how that perpetuates colonization unless that’s what you’re setting out to do with your methodology.
I am familiar with the regionality of language. I don’t understand your point, you’re simultaneously saying that you can’t have universal understanding, but we have gestures we instantly understand instantly so there’s no need to codify them, but they look different.
I think you’re wildly overestimating the scope of my proposal.
It would certainly be limited and rudimentary; I wouldn’t suggest a solution exists capable of any broad nuance. But gesture is a unique variety of communication, in that it can convey “innate” meaning in ways verbal language simply cannot, except in the case of onomatopoeia. Pointing is nearly universal, smiling is nearly universal, beckoning is nearly universal. Gesture is a spatial form of communication, centered around our primary means of material interaction with the world.
Grammar and syntax aside, I’d argue that it would be possible to assemble a vocabulary of universal concepts (eat, drink, sleep, travel, me, you, communicate, cooperate, come here, go away, etc). Certainly not a language for extended detailed conversation, but a codification and extension of gestures which are already nearly universal by virtue of their innate implications alone. Enough to communicate that you’re hungry, but not enough to send for takeout.
A universal language, at the level of any other sophisticated language, is obviously impossible. A formal codification of simple gestures to communicate at the most basic human concepts is much more doable.
I know this is a point of some contention among the deaf community, but how do you feel about the development of a “standard” international sign? Personally, and I’m speaking as a fully hearing person, I think a basic international sign should be developed and taught to everyone. Not only to facilitate communication with the hard of hearing, but also in loud environments and with those who don’t share a spoken language.
It’s my understanding that a large portion of the deaf community is hostile to the idea of a universal sign from a cultural perspective, since each regional sign has cultural content. However I think it’s a potential solution for numerous issues, with more pros than cons.
The woo-ey aspects are actually pretty interesting. Since the mechanism relies on focusing your subconscious, belief is crucial. If you don’t believe in your goal, and the efficacy of the method, your subconscious won’t buy-in, and without subconscious buy-in it flat out doesn’t work. Subconscious buy-in is the mechanism. You can’t try to consciously trick the subconscious, it’s in there with the one trying to trick it. You have to really believe.
A lot of people can’t believe that it’s internal. They don’t think that ability could possibly be in them anywhere, so in order to cultivate the requisite belief they have to attribute the mechanism to some kind of external woo. So even if the woo isn’t real, belief in the woo can be integral to the mechanism working.
That’s part of it, but not all. The world is a vast and complex place, you cannot possibly engage with, or even notice, the majority of the information available to your senses.
Your subconscious mind filters out information which isn’t significant to you, and draws attention to information which is. This is why when you get a car, it suddenly seems like everyone got that same make and model. That model didn’t become more popular, you just now have a reason to notice what was already there.
The Law of Attraction is one incarnation of the intentional exploitation of this psychological phenomenon. By attaching significance to some goal, and reinforcing that significance, you train your subconscious mind to notice opportunities in service of that goal.
I’d think it was funny if it didn’t read like another desperate attempt to stifle enthusiasm for progress and make the fascist guy seem more appealing, and why again? Because the president isn’t embroiled in daily antics? Dear heavens, not a boring President, quietly doing his job instead of tweeting scandalous things. Another boring, incremental legislative step in the right direction. The gears of democracy turning, yawn.
Why can’t milquetoast Biden be more exciting like the fascist guy? Why does the bureaucracy of running a country have to be comparable to tapioca instead of something flashier, with big banners and snazzy outfits and catchy slogans.
As stated, images are reasonable to blur, because having them in your screen would Not be Safe For Work. Text generally doesn’t really make sense to blur, especially an article published by a major news source. If it could be seen in print sitting on a lobby coffee table, it probably doesn’t qualify as NSFW.