The two countries are deeply economically tied, are political allies, share much of the power structure. I don’t think it makes sense to separate them in this context of morality, corruption level, or personal risk.

But it was an unconscious simplification that may not apply to others judgements. Apologies for the confusing inaccuracy.

Today! Very well news! Good journalisted news! Headlines:

My toddler says she will never eat food again in her life because we demanded she eats two pieces of broccoli.

The child rapist who shits himself says he will kill the world if they don’t do what he says.

My cat says (interpreted) that he is literally dying of hunger and sadness if he doesn’t get a double portion of his favorite food.

CEO of shoddily written autocomplete app claims it will kill us all if he doesn’t get a trillion dollars.

Fridge magnets fell on floor, say FOOD ANY IS IS SUN FOR THREE.

It is good to tell stories like this, for the public good, but I refuse to sympathize.

He chose to “make a fortune” in a famously and obviously through-and-through corrupt dictatorship known for figuratively grinding migrant labour to bonemeal to pave their roads. Literally chopping up journalists. Murdering villagers for a fake greenwashing project.

He put on tall boots and waded into this pool of blood to “make a fortune”. Smiling at people stabbing each other in the back, he brought his son along to grab soggy fistfulls of blood money. And he just… stayed there until his luck inevitably ran out.

The privacy statements are fucking lies.

I will not share my innermost mental issues with some group of 20-something “move fast and break things” sociopaths in Silicon Valley.

He’s really getting under my skin. I think I will go over there and give him a piece of my mind.

Appealing doesn’t get them out of paying.

These findings have been given an AI Doomerism PR spin.

The phrases “safeguards”, “deceiving” and “scheming” are incorrect.

The “safeguards” here are prompt begging, which is not in any way an adult’s attempt at a safeguard: https://simonwillison.net/2023/May/2/prompt-injection-explained/

The terms deceving and scheming indicate intent and agency that do not exist. I will count them as just plain lies.

The effect is that people imagine LLMs can get better by feeding their context windows with more rules, which not only makes it less likely the rule will be weighted significantly, but also causes the models to compress the now too-big context window lossily.

I’m going to work on my “Pyanodon Madness” Factorio save. I think I have to just give in and create properly engineered buffering stations for all the byproducts.

MCP and agents are hats-on-hats-on-LLMs, so you are not describing different types of AI here.

The use is evolving, towards lesser use. The drawbacks are becoming clearer and the significant subsidies are drying up.

I mean I could accept a diesel locomotive had a usecase in driving a whisk to whip cream for my latte, if someone else paid for it.

When I have to pay the cost myself… I’d just use an old fashioned luddite electric whisk and be left behind by all the super efficient diesel locomotive baristas.

Easy mnemonic to teach the month names to your kids!

Another, Year, Passes, Us, By, And, We, All, Die, In, The, End.

Yeah, you’d have to change the plug.

It is all going according to Trump’s plan actions and their inevitable and obvious consequences.

I am pretty sure I could do it sans phone and only the handle, by rapidly pulling the plug out of the socket and putting it back in.

Never thought to try it when I had the chance.

If presented with an old 1970-2000 era landline phone, I can call someone by rapidly hanging up in the pattern of their phone number.