March 26, 1893 – Plain Dealing, Louisiana (Plain Dealing High School): During an evening school dance, a fight broke out. Students fired shots, killing two immediately, fatally wounding two more, and injuring a professor (total: 4 killed, 1 wounded).

December 12, 1898 – Charleston, West Virginia: Young men (including students/former students in the context of a school exhibition) disrupted an event, leading to a brawl with gunfire. At least 6 killed (including students) and 4+ wounded in the chaos.

July 21, 1903 – Jackson, Kentucky (Cave Run School): Students James Barrett and Mack Howard dueled with pistols over a card game, killing each other; a 12-year-old bystander student was wounded (total: 2 killed, 1 wounded).

November 16, 1904 – Riverside, California (Indian School): A gunfight between pupils resulted in one student killed, another fatally wounded, and one wounded (total: 2 killed, 1 wounded).

October 8, 1950 – New Orleans, Louisiana (Booker T. Washington High School): Suspected gangsters (youths tied to students) fired on each other; 6 bystanders wounded.

May 5, 1956 – Seat Pleasant, Maryland (Maryland Park Junior High School): 15-year-old student Billy Ray Prevatte returned with a rifle after a reprimand and shot staff: 1 teacher killed, 2 injured (total: 3 victims).

October 17, 1961 – Denver, Colorado (Morey Junior High School): 14-year-old Tennyson Beard argued with a classmate, shot and wounded him, then fatally shot another student (total: 1 killed, 1–2 wounded).

October 5, 1966 – Grand Rapids, Minnesota (Grand Rapids High School): 15-year-old student David Black killed a school administrator and seriously wounded another student (total: 1 killed, 1 wounded).

Don't be so pessimistic, maybe they're just trying to give their buddy Musk and XAi a chance to catch up.

It's a terrible outcome for Amazon because it destroys Anthropic's revenue. Roughly half of Anthropic's customers are foreigners, and they wouldn't use Anthropic if its next generation model was banned while other providers' next generation models aren't. And if the US follows through and bans all Mythos-level models for foreigners, then in 6-12 months the entire global market will be overtaken by China when its models catch up, and Amazon will lose money on its investment in OpenAI too.

It is already possible: https://arxiv.org/abs/2603.08163 . You don't need to sync so frequently, so it can be done over normal internet, it's just less efficient (takes longer to converge).

The incredibly smart ones are able to use AI to multiply their productivity. The ones having a bad time with it from vibe coding and vague prompting aren't that.

Use Monero as much as possible. If enough people adopted it, there's absolutely nothing they could do to stop it short of turning off the internet entirely. Even China, with the strictest internet controls in the world, hasn't managed to stop people paying for banned goods and services in crypto there.

A human may learn and improve to avoid being fired, while Claude is incapable of that.

>Re the rest, it's just not my experience that models become incapable of making good decisions in cases where input token count > the context window, but ymmv based on domain.

If they've been trained a lot on your domain (maths, coding) then they can make good decisions. But I've just started using Mythos and even it makes some awful decisions in domains it's not trained on. Of course the majority of decisions are good, but it only takes a couple bad ones to sink a project.

All will perform roughly at the same level, just like today. It doesn't matter what provider you switch to, they'll all going to make mistakes because performing at a high human level requires far more business context and domain knowledge than is going to fit in even a few million tokens. Humans have incentive to learn and improve, LLMs lack even the ability to improve, as there's been pretty much zero progress on live learning and it's theoretically impossible for a fully-trained (saturated weights) LLM to learn new things without forgetting old things.

AI's having the opposite effect; it multiplies the productiveness of skilled software engineers while simultaneously multiplying the destructiveness of bad ones. The engineer who can shepherd a handful or Claude/Codex instances around simultaneously without producing slop will be immensely better compensated then the engineer who just gives vague instructions to the AI, goes to get a coffee and hopes for the best.

Having accountable people in key positions is a very important part of running a successful organization. Anthropic and OpenAI are never going to let you sue them when an AI employee makes a mistake; accountability is a strong moat.

In theory continuous learning (live weight updates) could help to some degree. But there's essentially no progress towards that because it requires solving a few hard, currently completely unsolved problems. 1. Weights drift over time and there's no way to re-merge them after a few tens of thousands of updates, so when a new model version was released there'd be no way to update existing continuously-learned models to that. 2. It'd allow permanent jailbreaking. And 3. A model can't learn new things without forgetting existing things, unlike humans brains which have hardware plasticity (like London taxi drivers having larger hippocampi due to having to memorize so many streets).

If this was true, Alphabet wouldn't currently be charging more for a worse product than OpenAI and Anthropic.