I mean, Leopold was literally fired from OpenAI after one year.

(Though I would hardly call Altman "a genius". He's a rather sociopathic VC guy who happened to be in the right place at the right time)

I've also had audio blocked in browsers since it was implemented.
Nothing more irritating than opening a page and suddenly some loud obnoxious noise blares out.

I'm holding out for so many things.

New fabs are finally being built but they take like 3-5 years, so that's not going to solve anything any time soon :P

Situational Awareness is a "hedge" fund that did the opposite of hedging - they made a bunch of leveraged bets that were all linked to each other in typical market movements. It was headed by Leopold Aschenbrenner, a guy just a couple years out of college, whose employment career had been very brief stints at the FTX crypto exchange firm (until it collapsed) and at OpenAI (until he was fired a year later over an alleged information leak). No financial management experience whatsoever. But he was into Effective Altruism, writing AI whitepapers, all that sort of stuff that Silicon Valley tech bros like, and so when he started a hedge fund, $45B was quickly pumped into it.

Nah, this isn't a sort of bubble-popping event. It's a tangential player in the stock market, not some key participant in operations or loans in the AI ecosystem. My main concern: inference IS profitable. Very. But the heavy leverage of the industry is also very real, and can very much still spread like a contagion. And a likely trigger for that is inflation, triggered by trade wars and the ongoing Hormuz and Ukraine conflicts (or worse, new ones added to the list!). One, inflation directly increases their costs, but two, it causes central banks to raise rates. This starves companies of capital (both loans and equity), incl. to refinance existing liabilities, while also hindering income (e.g. new orders get put off or cancelled). It can easily flip a leveraged company into insolvency, and then that can ripple if there's nothing to stop it.

An investment opportunity can be brilliant, a massive world-changing field with huge margins, but still be a terrible investment. One, because the companies you invest in need to survive continuously and not get blipped out by an adverse market event until they have more ability to withstand them. And two, also, because there's a problem where investors don't merely value the whole market as if it has high odds of success, but value the specific player they're invested in with high confidence as if it will dominate said market. Which, obviously, all players combined cannot do.

Whether it's profit or not depends on whether people buy anyway. I've decided "no upgrade this year".

First, searches of their persons and effects does not mean reading and making a copy of their letters and books. It means looking in the trunk and verifying that it is just letters and books.

Second, there were no border agents in 1789. You could walk or swim into the U.S. from Canada and Mexico freely for (statistically) most our nation's history. The first border patrols began in the 1920s to restrict immigration from certain countries. There were laws in the late 1700s regarding import of *cargo*, and that was handled by the folks who ran each individual port, but again, that's not the same thing as searching a passenger's person and making copies of their letters. Those laws applied only to identifying goods that were likely intended for sale, for tax policy/import duty reasons.

This bastardization of border policy to include searching the contents of people's papers and devices is a late 1900s perversion of the Constitution.

So no, pretty much everything you said is wrong.

It still is if you do it yourself and don't have ChatGPT do it for you. But it isn't a useful indication for parents or teachers, because the kids can just have ChatGPT do it for them, and moreover are encouraged to do so by the pressure to get good grades. By making the homework be graded only on adequate participation (which may be per-student) rather than on correctness, you shift the focus back to the homework being a way for students to get ready for the test and make it clear that having an AI do the work will prevent the teacher from knowing how well the student is doing, and will likely result in a lower grade on the test. Once the motivating factor of grades is eliminated and, in fact, reversed, they'll stop doing it, or will use it only to help them understand or to check their work, rather than using it as a crutch to actually do the work.

Any child gifted with high motivation and broad interests wants to practice stuff they already understand less so that they can do other things and learn other things. A child's time is a zero-sum game. The more time they spend doing things that are less useful, the less time they have for doing things that are more useful.

The ones who mindlessly work more problems because they are motivated to get good grades are the ones who aren't doing band, orchestra, chorus, or theater, who aren't going on field trips to learn about the world, who aren't getting the breadth of education that will prepare them for the real world, and will mostly end up in boring (but probably reasonably well-paying) jobs for the rest of their lives. They won't ever be the rockstar in their field, but they probably won't fail horribly. And I guess that's okay, but IMO it's not what we should try to inspire kids to become. The motivation should be to learn for learning's sake, and excess repetition is the opposite of that.

Meh. If that's your experience, then there's no hope for improving anything, because the kids who don't do well are going to pass anyway. This is how we end up with a bunch of graduates who can't actually do anything. I'm not saying you're wrong, but if you're right, then it's critical to fix that at the same time.

They're overlapping, but not the same group. Some smart kids are hard-working and eager to learn. Some are rule-following. Some are both. Some are neither. What I think you meant is that they are not dependent variables.

The effect of this would be that the teacher would quickly intervene and tell the kid that they need to do more of the work, because they don't seem to have fully grasped the concepts. If the kids ignore this, then at the parent-teacher conference after the first six-weeks period, the teacher would encourage the parents to either make their kids do more work or put their kids in an after-school program that would do so.

And if that doesn't work, you can also make the participation points be contingent on whether they are grasping the concepts, where if it is clear that they're not getting it and aren't doing enough work to learn it, you tell them that they won't get participation points unless they do more of the problems.

You left out informative, but otherwise I pretty much agree with you. The problem is that I expect MANY jobs are going to evaporate. Whether enough new jobs will appear (that are easy to cross-train into) is another question, and I really find it dubious.

We won't find out the answer this year, but we may find out next year.

You may think it sounds like gravitas, but to me is sounds like a desperate ad man.

The methods the summary implies that they use are quite unreliable on an individual basis, but probably pretty good on a population level. So talking about the increase in frequency is probably valid.

Certainly! Your point raises a crucial, multifaceted issue that merits deeper examination. It's not just about detection accuracy, it's about the evolving interplay between human intention and algorithmic assistance — a dynamic that resists simple binary classification.

When we delve into the mechanics of modern AI tools, it becomes evident that the boundary between "correction" and "generation" is increasingly porous. Consider the following nuances:

Gradual authorship spectra: Text exists on a continuum — from fully human-crafted prose, to human writing with AI-suggested punctuation, to passages where the AI has restructured entire paragraphs. Detection tools, however, often operate on categorical assumptions that fail to capture this complexity.

The feedback loop phenomenon: Furthermore, as writers internalize AI-suggested patterns — Oxford commas, em dashes, and certain syntactic rhythms — their "human" baseline shifts. This creates a fascinating, almost paradoxical scenario: the more one learns from AI feedback, the more one's unassisted writing may trigger detection algorithms.

Contextual markers vs. stylistic choices: Notably, certain signals — like the increased use of "delve," "testament," or negative parallelism — may reflect genuine stylistic evolution rather than automated generation. It's a testament to how AI influences language culture broadly, not just individual documents.

In conclusion, the challenge you identify is not merely technical but philosophical — forcing us to reconsider what we mean by "authentic" voice in an era of human-AI collaboration. The question, ultimately, is not whether a tool assisted in the writing, but whether the ideas, arguments, and intellectual labor remain genuinely human.

I hope proves helpful! Please let me know if you would like to delve deeper into any specific aspect of these findings, or if you would like me to explore the broader implications for content strategy.

"do a little research" - you do realize that your post would have had just the same educational value but not made you come across as a jerk if you had just left off that first sentence?

This is essentially a high-stakes gamble. The payout is potentially millions, so they just try whatever. And I bet if it all turns against them and they're forced to pay legal costs etc. they'll just file for bancruptsy.