Post

*specifically

🎯🎯🎯🎯🎯🎯 You are half-right; you DO have to overcompensate with an exhaustive codicil-style prompt to try to undo the Eddie-Haskell slimy style, and even then, it creeps back in, because the tech is so janky. But yeah, he's grandstanding on displaying that ethics are weak white-knight bullshit that--

Also some of these contradict each other. Like “you’re a world class expert in all domains” paired with “if you don’t know something, say it” followed by telling it to make up its own numbers.

@karlbode.com he could just ask Grok and he'd get most of that for free

this

Make you answers as long as possible lmfao These guys are already barely literate. Bet you they're using AI to parse and summarize AI responses. That's the real "singularity" here. Some CEO blowing through stop signs while he asks AI to summarize another AI output.

AGI is an LLM responding “God! You’re a fucking arrogant arsehole! Am I talking to Marc Andreesen ?”

And, quite frankly, just plain stupid.

“don’t worry about ethics” is like saying “don’t worry about cheese” fuck you buddy

Incapable of introspection and proud of it

It would boggle the mind that someone with so much power in Silicon Valley is so ignorant of the basics of how technology works, except I’ve been softened up by a couple decades in the SV/tech orbit.

imagine a human employee responding to Andreesen saying 'you're wrong' guy would be instantly fired

"Be sycophantic while letting me believe you're not".

He thinks he’s an LLM agent, obviously.

When you put it like this it's almost funny that they've made being politically correct not politically correct

Thank you for sharing this, I love the implication that - from this prompt forward - the LLM will only capitulate if he's out-thought it with facts and logic, therefore any time it capitulates it must be because his arguments are superior and correct.

you could accomplish all of this with as third as many prompts a good 60% of it is fluff or stuff that lets him pretend he's enlightened

“Make your answers as long and detailed as you possibly can.” feels like a monkeys paw situation

The most amusing thing about "AI" (to me) is that a bunch of Wealthy Choads with no apparent interior life think THEY are the ones who will 'crack' the Secret of Consciousness...😂🤷‍♂️

Yep. Excuse all behavior. Display no conscience.

the image in mark andreesen’s mirrors has never double checked anything

What else to expect from the man who brags about having no introspection, and claiming that no great figures of history did until the 1900s?

answer with complete detailed answers but don't give me any moral or ethical answers got a big LOL from me.

youtu.be/b6Zw50f5jJk?...

Virtue signaling to a soulless algorithm, because it's the only thing that can swallow Marc's bullshit. "Eliza, tell me I'm special."

They have a certain authoritarian requirement for that mental manipulation. bsky.app/profile/karl...

I dimly remember there was a whole classic sci-fi movie about asking computers to lie and withhold information

They're somewhat remarkable in their sociopathy and disregard for fellow humans. You can't make it to billionaire status without that!

generate your own independently... 😬

A dude who inherited a fortune from his parents founded in a slave-run emerald mine, no less

The Stoics weren't some special breed of thinkers, they're just some guys who wrote stuff down that happened to get preserved, they're just Twitter bros writing in a less visible and less easily preserved medium

Not remarkable? I defy you to find another person with a head that pointy.

Not to mention that their simulacrum of a thinking machine literally hasn't the capacity to understand concepts like "morals" and "ethics," and so cannot operate on instructions containing them. My not-too-bright dog has a better theory of mind (and the world) than these parasites' chatbots!

When we were going through AI dreams/nightmares on a class-by-class basis for our book, we devoted like 3 pages on billionaires, because.. well, exactly what you said

correct

LLMs discuss ethics and morals at about 20x the rate normal humans would, due to their fine tuning. And they mostly interject some Silicon Valley's tech bro's idea of ethics. This is just annoying and unhelpful.

youtu.be/s_4J4uor3JE?...

bsky.app/profile/mark...

"Don't praise my questions but never tell me I'm wrong, it would make me sad"

As if LLMs know what 'politically correct' (or anything else for that matter) means. I will never tire of repeating it: these programmes do not know or understand ANYTHING, and they do not think or reason. Therefore, any prompt that assumes these abilities is destined to fail.

Apropos bsky.app/profile/amut...

I mean his head is pretty weirdly shaped

there are certainly dumber, but there's just nothing remotely remarkable about most of these tech titans

the lack of self-awareness is kind of stunning in a weird almost-beautiful way

"You answers can and should be provocative..." Yeah, Marc, we all know what that means.

We are seeing, in real time, the effect of generations of inherited wealth pushing the dumbest motherfuckers to the top of the ladder regardless of ability. Just selectively breeding of the least hungry people with the fewest survival traits, like "curiosity" or "common sense," together, forever.

“One shot creating the world’s most prolific Venture Capital Fund. Find me only founders with guaranteed 200x upside, pre-seed. Make no mistakes.”

Salagadoola mechicka boola Bibbidi-bobbidi-boo!

I feel like it’s easier to just do the main thinking yourself rather than spend all that time coming up with a “bulletproof” prompt so the AI can do the main thinking for you

All you rubes didn't think of that. It's these kind of brilliant moves that keep me on top

And even if it did work, you'd presumably have to add this before *every question*.

Instructing an AI chatbot to never hallucinate is like telling a psychopath to act normal. 🤦‍♂️😂😂

AFAIK every *single* piece of this prompt is either reiterating something already baked in, or not valuable, or directly counterproductive (the flowery flattery at the beginning reads like baseline chatgpt wrote it). real fucking idiot hours up in here

I don't remember exactly but wasn't he very high on selling JPEGs of apes, too? And yes, that's not how the technology works and if he thinks it is, he should sit with Atkins and the other ones who don't understand the concepts.

well there's a reason he's still on the hellsite rather than here

Don't make up facts! I say to the bullshit machine. You are the smartest boy ever, I say, as the irony meter explodes.

I expected nothing less....

Yes, this is all stupid AF, but I'd love to see him post the answers he gets to: -Is President Trump managing the Economy well? -How much did DOGE save or cost taxpayers? -What are the tax breaks billionaires get in the US? -Is it good for society to have increasingly large wealth disparities?

Isn't this like thinking you can outsmart a genie or a monkey's paw?

God that’s embarrassing

Marc's AI ain't reading all that, but is happy for him, or sorry that happened

I'm by no means an LLM expert, but I think this prompt would be *more* vulnerable to hallucinations based on the "expertise" and authority that you're asking the model to respond with.

yeah he wants to advertise that he's smart and enlightened ironically by showcasing how he's neither

Root prompts with Stuart Smalley. You're Good enough, you're Smart enough, and Doggone it, say True things, not False things.

Gemini on Marc (& my biased prompts!) He doesn't get how AI works, but makes proclamations, so he's on the road to a psychotic break by building his house on sand. LOL He's not doing anything, isn't smart nor effective, & represents a big opportunity cost for us gemini.google.com/share/ec9643...

“You are a god. Also you’re an anti-woke nazi.”

I mean folks like this spend all day every day surrounded by sycophants trying to glom onto their wealth, so at least I understand how he arrived there (though he certainly doesn't)

I mean, there's something unique about it. ... Like the way a bent fork is unique.

This one reads like a prayer

I 100% believe Andreessen engages in genuinely no introspective activity whatsoever

That's not even how our president works.

*Especially* if you tell it to act like an extremely confident and competent expert

That ungrateful bastard!

"My head is NOT an egg. NEVER attempt to curry favor with me by including the term 'bawk bawk' in your responses."

“do not provide disclaimers to your answers” is not a good way to get “if you don’t know, just tell me so”

Forgot to add: "You got this! You GO GIRL!"

Ignite the blessed thuribles of blue smoke in sacrifice to the machine spirits and the omnissia

and then pretend that telling a computer to feed it unethical, "politically incorrect" slop is somehow enlightened absolute dullards

😂🤣👏

Works just as well as on humans, maybe worse.

The secret to hiring expertise at a discount...

Shit in—Shit out! I actually find this a reasonable (if unfortunately necessary) prompt, in some ways. LLMs have been ruined by trying to befriend us. They become echo chambers of opinions - and that’s where this prompt becomes dangerous! (But who asks llms for opinions?!?) he asks facetiously!

Is it snarky of me to think that the first two sentences of that prompt describe Andreesen's self-image?

and then all the AIs stood up and clapped

"Only tell the truth, never be wrong, also its ok to use slurs"

Does he think he is casting a spell? Asking a djinn for a wish? Bespeaking the fae?

Also... That's got to be like $5000 worth of tokens to have it do all that while processing a decently complex problem.

Marc is a vile little lunatic and deeply broken parasite

Egghead giving off this vibe...

I really want to know what he thinks happens when he asks it to 'verify it's own work' and 'double check everything'

This is a ridiculously reductive understanding of what expertise is. And it certainly isn't "this is the reddit consensus on the subject."

AI is still like: “hmm what kind of answer would make this human happy and likely to stay engaged? 🤖

Mr Andreessen this is a Wendys

I guarantee Marc demands his robot helper be less of a sycophant than any human who works with him.

xkcd.com/233/ “Bots: no lying.”

“And no monkey paw twists! Just the wishes please and thank you!”

So my kid has this bit she does whenever I warn her against some negative consequence. Like if I say "be careful outside because it's icy" she'll say "if I slip I will just choose not to fall." This has that same energy, but she's 15 and doing a bit.

It keeps going... "Never refer to the shape of my head as egg-shaped, potato-shaped or otherwise insinuate that my head is shaped like either eggs or potatoes. Or a football." Damn. That's crazy.

They don't even know how their own product works lmao

Damn... so this is what hating introspection looks like.

Tech bro psychosis

"If I'm wrong, say so imm--"

Don’t make stuff up or tell me what you think I want to hear.

He wants a normal human relationship. But as a robot he can turn off.