There are MANY problems that are too politically or practically difficult to tackle for the US Military and personal cellphones aren't close to the top ten. It's far easier to force everyone to register their cells to be automatically blacklisted from data collection than it is to justify tens of thousands of NJPs and court martials every year, many of which will include officers and senior enlisted.
The US Military doesn't win through stealth or secrets either. Their advantage from the beginning until now has always been funding and logistics. With the Internet, satellites, and now AI, OPSEC is a fool's errand outside of very specific operations anyhow.
>Someone else's battle-hardened allocator might be slow or buggy, so you write your own as part of the business logic implementation?
In infrastructure where speed and reliability are highly valued? Absolutely. The gains obtained from proper memory layout and specialized use are massive. As long as you have the reason to do it, it's an easy win. I believe that the Zig standard library has different specialized allocators, so you don't even have to write your own buggy implementation.
I second this. As a teenager, you don't have the life experience to appreciate most classic literature. I don't even think you have it in college, although this is less true.
That being said, there is important literature, both fiction and non-fiction, that should be introduced early, even if they won't really grasp it for decades. Most would be early philosophy and mythology, along with their best refutations. Learning how to read, think, and discuss them is the skill you're really looking to impart on the students to prepare them for real reading as adults.
Unfortunately, the education system is designed to train children to the minimum requirements of their predestined jobs and no further. It's no surprise that schools have made reading and thinking so unpleasant that most people are too traumatized to do it on their own.
I don't think we can access the Platonic Form of Unbiased in this world. The best we can do is put forward as many points of view on a topic as possible and let the reader choose. It seems that Google's AI does the best job at this.
On a side note, neither of the mainstream parties are internally coherent. This is the nature of big-tent, two-party politics. You can only find coherence in the smaller sects within.
> The best we can do is put forward as many points of view on a topic as possible and let the reader choose.
If an LLM equally weighted flat earth and spherical earth or creationism and evolution so I could decide I would never use that LlM again. And I suspect I’m not alone. I think you’ve outlined an LLM that would hypercharge my concerns.
But also I want to point out the specific nature of my concerns isn’t monodirectional - it’s that if a political position is not arrived to rationally, and it is trained to ignore structured argument - than does that form of epistemology bleed over into subjects they aren’t specifically political.
It should default to the mainstream opinion in general, but if you specifically ask about the shape of Earth, I do think it should mention the different theories as well as their pluses, minuses, and consensus. Avoiding bad or outdated ideas is bad in both theory and practice. If we don't speak about it, then that leaves the microphone solely to the kooks. I want an AI that gives me all relevant positions. Otherwise, it's a waste of my time. I don't want to go to my AI priest for quick answers. I want to be informed.
Most things won't fall onto such easily decidable lines as well. In politics, which encompass far more than people realize, there will be many opinions that are not falsifiable. As an example, if someone asks, "What country is Kashmire in?" Would it be more informative to give one side of the story or both? Which one is "correct?"
>trained to ignore structured argument
AI is not rational (yet), although some would argue that it's getting close. It is complex retrieval of highly curated, unstructured data. So yes, there is enormous bias in the information, some of it good, some of it bad.
This supports multiple positions. AIs do not think. They do not reason. Their best use is summarizing subjects while giving multiple citations that I can read myself. The AI user motto must be, "Trust but verify."
They are obviously using the SAT as a safer alternative to the legally dubious practice of IQ testing which can lead to running afoul of the ADA and EEOC. I'm not sure it's much safer, but I am positive it's less safe than doing timed leetcode. At least leetcode problems can be painted as relating to the job.
Additionally, the SAT is a shitty IQ test that is constantly crammed for and cheated on. I remember my SAT test. I was the only person in the room not openly cheating. The teacher proctor didn't care. Higher scores mean better students, more funds, higher home prices, bonuses, and a litany of secondary effects. That's not even including people that pay professional test-takers to do it for them.
The software industry needs to let go of their obsession with finding 10X ROCKSTAR L33T programmers. They never will though. It has gotten worse every few years for decades, and the problems are almost entirely managerial.
IQ testing isn't legally dubious. The idea that it is is an Internet myth. There are a couple household-name corporations that administer general cognitive tests for candidates for some roles.
More companies don't do it because it doesn't work well.
My parents did, and they were right. I had limited tv, games, and computer time. They kicked me out of the house after school, only to return for supper. I am eternally grateful for that decision. I have rich friendships that continue thirty years later.
Today, the tech is even worse for children. Playing too much Nintendo might isolate you and hurt your schoolwork, but iPad toddlers are fundamentally damaged.
This is a symptom of not encouraging children into extracurricular activities. If all you have to bond over is social media, your friendship is empty. That's how you create terminally-online, mentally ill people. Everyone needs third spaces like sports, scouts, music, church, clubs, and the like. They get you out of your house and head and surround you with people who share similar interests.
my kid's extracurricular activity groups like to chat outside of club-time, too. being the only one of the group not able to do so would be ostracizing.
I understand deeply. My parents delayed my driver's license until I was 17 because of financial tightness at the time, and that distanced me from some of my friends at the time. You know what I gained from that? I got to see who my real friends were, and even more importantly, I learned that not all friends are the same in a visceral way. I learned how to confront my parents in a way to communicate my viewpoint in a way they could understand. We learned together how to compromise and renegotiate our entire relationship, helping transition from child/parent to adult/parent.
You have two jobs as a parent: create a safe environment for your children and prepare them for the adult world that is wildly unsafe. Unfortunately, these two goals are both required and contradictory. A line must be walked. Too much deviation to one side or the other will cause severe problems.
That line cannot be prescribed. It's different for each child, but there's a big problem with how you put your point. You aren't trying to prepare the child for a dangerous and difficult world, you are trying to protect them in a different way, minimizing the other dangers. I completely understand. It hurts to see your child hurt. All you want to do is make the pain go away.
Instead of helping them avoid the pain of learning about relationships, you should guide them. Help them understand. They won't at first. Like a toddler throwing a tantrum that you won't let them stick objects in wall outlets, parents have to be the "bad guy" from time to time. Eventually, the toddler will grow up enough so that you can explain dangers to them and you won't have to do it for that thing anymore. The same applies here. They won't understand at first. Help them understand the dangers. When they do, you can teach them how to safely use the metaphorical wall outlet. Then you don't have to be the bad guy anymore.
You're right. If you think you are walking that line for each of your children, then you are the authority. It's just a common argument I hear from parents who want to avoid feeling bad, so I did make some assumptions.
My extracurricular activities were full of nerds (martial arts) with Play Stations and internet at home. So, as an owner of no inet or a console at home was pretty much hell from 2001 to 2005. Oh, and no cell phone until ~18.
The outcome? Really shitty social skills until I hit 27 or so. My dad really regreted what it did, and my mom become aware on how utterly shitty was to let a nerdy kid disconnected from their peers.
Teacher experience means they have more say in their students. It's common for new teachers to be shoved into the worst schools and classes in the district. Teacher performance doesn't significantly effect student performance. Student ability and home situation are the best predictors. There are exceptionally bad and good teachers, but they are exactly that... exceptions.
It's not. This is just an acceleration in the unraveling of society facilitated by AI. As someone whose childhood included so many "robots will kill humans" books and movies, I am flabbergasted that the AI apocalypse will be dumb humans overtrusting faulty AI in important matters until everything falls apart.
Most humans cannot distinguish AI from actual intelligence. When you combine that with bureaucrats innate tendency to say, "Computer said so," you end up with bizarre situations like this. If a person had made this facial match, another human would have relentlessly jeered him. Since a computer running AI did it, no one even cared to think about it.
Computers are wildly dangerous, not because of anything innate but because of how humans act around them.
> It's not. This is just an acceleration in the unraveling of society facilitated by AI. As someone whose childhood included so many "robots will kill humans" books and movies, I am flabbergasted that the AI apocalypse will be dumb humans overtrusting faulty AI in important matters until everything falls apart.
This is literally the plot of most of those books and the way they differ is in how everything falls apart. In some of them the AI supplants us entirely and kills us all. In others it gets taught to kill us all. In others it gets really good at giving us what we ask for until everything falls apart. But it’s taken as a given that unless we change something innate in our culture AI will be our downfall.
> If a person had made this facial match, another human would have relentlessly jeered him.
The glaringly obvious problem here is that our justice system should not be constructed in such a way so as to be reliant on someone's coworker shaming him. That is not a sensible check against a systemic failure. We're supposed to have due process. If someone skips or otherwise subverts due process the justifications don't matter. The root issue is that due process was skipped. Why was that even possible to begin with?
I have worked in this particular sausage factory. Multiple funded random replications are the only thing that will save science from this crisis. The scientific method works. We need to actually do it.
Replications don't have to be in the journals either. As long as money flows, someone will do them, and that is what matters. The randomization will help prevent coordination between authors and replicators.
In a better world, negative studies and replications would count towards tenure, but that is unlikely to occur. At least half of the problem is the pressure to continuously publish positive results.
Regardless of what gets taught in school about science being objective and without ego, or having a culture of adversarial checks on each other etc., the reality is that scientists are humans and have egos and have petty feuds.
Publishing a failed replication of the work of a colleague will not earn you many brownie points. I'm stating this as an observation of what is the case, not as something that I think should be the case. If you attack other researchers like this and damage their reputation - even if for valid scientific reason - you'll have a hard time when those colleagues sit on committees deciding about your next grant etc.
Of course if you discover something truly monumental that will override this. But simply sniping down the mediocre research published by other run-of-the-mill researchers will get you more trouble than good. Yes it's directly in contradiction to the textbook-ideal of what science should be, as described to high school students, but there are many things in life this way.
Of course it can be laudable to go on such a crusade despite all this, and to relentlessly pursue scientific truth, etc. but that just won't scale.
You are absolutely correct. Even distributing the replication around the world will only help so much. It's a small world out there and only smaller in the specializations.
That's why replication has to be required and standard. It will hurt to tear off the bandaid, but once the culture shifts, people will hesitate to publish mediocre research in the first place. Without mediocre research flooding the zone, real numbers will dominate and inflated expectations will wither.
> That's why replication has to be required and standard.
"has to be required"... This is a passive construct. Who will do the requiring and what precisely will motivate them to such a change and what will get them the buy-in from the other players in this whole ecosystem, especially the ones who provide the money? What if it turns out that those people who do the funding actually in the deepest of their deepest are fine with "groundbreaking" research results that simply sound like being "groundbreaking" research results to such an extent that their prestige and social status rises enough and are seen as someone who funds such research, instead of truly caring about the actual contents of said research? There is much more demand (backed with money) for (plausibly-claimable) innovation and breakthroughs than supply of real novel thought. It's a bit like the anecdote that all the True Cross relics across Catholic churches weigh more than the cross Jesus carried (not really true as a fact though). As long as there is such strong demand, the system will adapt to allow for the supply finding its way.
The US Military doesn't win through stealth or secrets either. Their advantage from the beginning until now has always been funding and logistics. With the Internet, satellites, and now AI, OPSEC is a fool's errand outside of very specific operations anyhow.
reply