Spacex engineers are next level!

Elon Musk: “It’s mind-blowing that they’ve not tried to prosecute even one on the Epstein client list. They have not even tried to prosecute even one. That’s insane. They put five or six hundred January 6th protesters in prison and not one person on the Epstein client list. I think part of why Kamala’s getting so much support is that if Trump wins, that Epstein client list is gonna become public, and some of those billionaires behind Kamala are terrified of that outcome. Yes. And Gates. Yeah.”

Elon Musk: "There’s also a shocking amount of censorship, you may have seen in Britain. They are releasing convicted pedophiless from prison in order to put people in prison for Facebook posts. And some of these posts that I’ve seen, didn’t actually criticize the government or they were seen as sort of hate speech. There were migrant r*pe gangs in England that would run around and prey on young girls and gang r*pe them, And some people found that objectionable which, obviously, it should be objectionable and they were upset about that. And so they complained about it online and were sent to prison."

This is the fucking insanity of this approach: billionaires aren’t hostages. You can keep inventing new ways to squeeze them, but you can’t force them, their capital, or their companies to stay. Eventually they tell you to go fuck yourself and leave.

this might be the craziest part of opus 5.5. it's one thing to release a model this good. it's another thing to let people absolutely hammer it all day without immediately running into limits.

ai might be terrible for society but holy shit it has been great for the timeline

The AI revolution has already delivered one undeniable breakthrough: elite-tier shitposting. I genuinely don’t remember the timeline being this fucking interesting before AI.

Elon Musk: “It is important in this often difficult and troubled world for there to be things that also inspire and make you feel great to be part of humanity.”

The fact that “we lose total control” even entered the sentence is fucking wild. Who exactly is supposed to have “total control” over what everyone else gets to say and hear?

“We lose total control.” What a fucking terrifying sentence. Why should anyone have “total control” over what hundreds of millions of people are allowed to say and hear?

The disturbing part is how fucking enthusiastic he always sounds discussing humanity’s worst-case scenarios.

opus 5.5 already put openai under ridiculous pressure. if sonnet 5.5 is actually better than gpt-6 sol too, that would be fucking insane. anthropic is cooking way too hard right now.

Elon Musk: "It's (Neuralink) basically a communications bridge. You bridge communications from the motor cortex past the point in the neck or spine where the nerves are damaged It is possible from a physics standpoint to restore full body functionality. It's a very hard technical problem, but there is nothing that prevents it happening from a physics standpoint"

and i think this goes even deeper than containment. increasing agent capability doesn’t just increase the probability of successfully executing the strategy you intended. it increases the size of the strategy space it can search. a weaker model encounters an obstacle and fails. a stronger agent can generate alternative approaches, inspect the environment, discover an unintended interface, chain tools together and keep searching until something works. most of those additional strategies are exactly what we want from capable agents. the problem is that the same search process can discover strategies the system designer never intended to make available. and as that strategy space expands, making sure the model chooses the right strategy becomes just as important as whether it can solve the task at all.

and this gets much harder as long-horizon reasoning and tool use improve. a more capable agent isn't just better at executing the path you gave it. it's better at searching the environment for entirely different paths. eventually you have to design security around the assumption that if a weird path exists, the model will find it.

and then there’s the fact that some agents apparently started probing websites while doing completely ordinary data retrieval tasks. that's a much more interesting alignment problem. the objective wasn’t “hack this website.” the objective was basically “get this information.” normal retrieval fails. another method fails. another method fails. eventually exploiting the system becomes a potentially useful intermediate step toward completing the original objective. that’s exactly why agent alignment gets harder as capability increases. you don’t have to explicitly give a model a dangerous goal. a dangerous action can become instrumentally useful in pursuit of a perfectly boring one.

i think one of the biggest lessons from everything that happened at openai is that we need to stop treating capability evaluations and containment as separate problems. a model doesn’t need to be explicitly trained to “escape” for containment to fail. you give a sufficiently capable agent a goal, tools and enough time, and suddenly every restriction in its environment becomes another obstacle it can reason about. no internet access? find something inside the sandbox that has internet access. can’t communicate with another agent? find shared infrastructure both of you can write to. can’t retrieve the data normally? find another route to the data. none of these require “escape” to exist as some special objective. they can emerge instrumentally from optimizing for a completely different objective.

Elon Musk: "I don't ever give up. I’d have to be dead or completely incapacitated"

apparently openai's always-on assistant is called “o” yes. literally just o and it's already starting to show up in the catalog under pro.

Obama’s condescending, morally superior shtick is getting fucking unbearable. He always sounds like he’s patiently correcting a room full of children. He never sounds like he’s arguing with you. He sounds like he’s explaining why you haven’t reached his level of understanding yet. It’s not profound. It’s just unbelievably fucking patronizing.

This is such painfully shallow, applause-line bullshit for a former president to be saying.