zmgsabst

Scorpions had a turn when the giant ones preyed on our lungfish ancestors.

This is our turn; eg, similarly dinosaurs preying on proto-mammals and our abuses of avians.

You absolutely can call them and make it clear that their playing games is causing you reputation damage ahead of an IPO — so you need that fixed.

A large conglomerate isn’t interested in pissing off a client because their subcontractor has a middle-manager playing fuck-fuck games with a union. That’s exactly the sort of business situation that executive communication can fix.

Is the claim “created”?

My understanding is that the CIA influenced the rise of modern art over Soviet realism, eg, by hosting shows and bidding up prices.

That is “created” in the sense that a movement would have been smaller or fizzled without their patronage, but not “created” as in invented absent an underlying cohort of sincere artists.

If you believe the CIA was an important and necessary part of the movement, you seem to agree they created.1 modern art, even if they co-opted rather than created.2 the style.

Circa 2015 when the reformulation of meth made it more likely to induce psychosis — as the new flow chemistry version altered which isomers were made.

At least, in Seattle.

I’m working on a whitepaper on a layover.

> I know good sentences. I know the style of prose I like to read. I know how things can veer between florid and poetic and where the line is. I know how a sentence can be completely different if you read it as though it was spoken, instead of to yourself (some of mine are written to be read.)

Your perspective doesn’t really speak to me at all — because I’m writing as the way to explore and articulate ideas, while I’d get rid of the words as an unnecessary mess if I could. (And do when I can: diagrams, math, etc.)

This is partly why I don’t worry about “LLM slop” — we had plenty of artisanal human slop before.

You can just reverse the two phrases:

“I felt X, Y, Z, so I did P, Q, R, but I’m sorry for that because it caused you A, B, C.”

Your usage of “but” still invalidates the preceding phrase on behalf of the latter, which works here because the affirmed phrase is the apology.

I don’t think there’s anything wrong with stating your side — but if you mean it, end on the apology.

Reasoning is mechanical.

I think a problem is that previously thought, reasoning, and intelligence were always co-occurring, but now we have machines capable of (limited) reasoning that do not think or have intelligence.

“Aha” as a single token records the LLM discovered it made a mistake and needs to pivot.

On the next forward pass: it rediscovers the mistake, its “aha” noting that, and then provides the first token of the new idea.

That “aha” contains information: the previous conclusion was somehow insufficient.

I know almost nothing about music; can’t play a bit.

I can still distinctly remember the first time I heard Rachmaninoff performed in person — and the shock at the “brutal genius” of the art.

I think sometimes art just speaks to you.

Politely? — yes, all the time.

A frequent first question when a subordinate asks me for help:

“What have you found so far, searching for answers?”

People all agree on that sentiment.

Where they disagree is if you add value on the knowledge that makes the prompt possible or add value in rewriting AI to pretend you wrote it.

I agree with you about which objects are motive, ie, LLMs do just sit there unprompted.

> I believe that this recurrence is where "intelligence" lives - and I believe it is the difference between a thinking being and a stochastic parrot.

My objection was to this, on technical grounds: LLMs exhibit intelligence.

1. They reason in an internal type theory.

2. This type theory is meaningfully encoded from the actual data and not stochastic, eg, research on language geometry.

3. Intelligent and reasoning doesn’t entail self-motive; that’s merely a spurious correlation from the fact that until now, we’ve only known intelligence animals.

You cannot conclude something is merely a stochastic parrot because it isn’t self-motive.

Project LANA ran into the same roadblock as Scholze and Stix, when they attempted to formalize the proof in Lean.

I’m not convinced.

In history, we made much more use of hitting things with bows than abstractly comprehending arrow flight.

Abstract polynomial algorithms are of no use either, eg, translation may require galactic constants or high powers that are still intractable.

We only compute with two kinds of things:

- small data; or,

- extremely lower power and coefficient algorithms

We lack the power to, eg, use a quintic algorithm in anything but nearly trivial cases.

To be fair, “s’kiddies will exploit low hanging fruit with semi-automated vuln scans” is a lot more realistic threat than “LLMs are going full Skynet any day now!”

I’d definitely suggest companies start addressing that first concern, even if I’m in the camp that thinks the second is fantasy.

> It's an LLM, it doesn't think. It's a machine that predicts the next token, given a sequence of tokens.

These can both be true, particularly when there is substantial state associated with each token prediction.

LLMs routinely mindlessly repeated propaganda in their training data.

I’ve yet to see a single one identify that on its own, rather than only reluctantly when confronted when documented facts it was objectively wrong.

So no, I doubt LLMs will do anything but promoted propaganda.

Quotas were rules illegal in 1978. I listed them as one of a sequence of continued (and later ruled illegal) bigoted practices, over time.

> 1978 — Regents of the University of California v. Bakke, 438 U.S. 265 (1978) — The UC Davis School of Medicine admissions program violated the Equal Protection Clause with the institution of quotas for underrepresented minorities. However, Justice Lewis F. Powell Jr.'s concurring opinion deemed diversity in higher education a "compelling interest" and held that race could be one of the factors in university admissions.

https://en.wikipedia.org/wiki/Affirmative_action_in_the_Unit...

This is how language evolves.

For supermajority: the construction means “beyond majority” or “above majority”, it is implemented as a higher percentage in most systems, and there is a utility in being able to refer to high ratio majority.

So people have repurposed a technical word to a general word, of roughly the same meaning.

Even if the total cost is the same, if our payments better align with the behavior we want to incentivize, we may gain greater utility from the spending.

So it can matter how we distribute that revenue as to whether or not the business responds in the desired way, eg, actually investing in new capacity by linking payments directly to new capacity.

They refused to give money unless your application included a section explaining how you supported Democrat social ideology, during the Biden administration.

So called DIE-statements.

Sure it is — see the Democrats refusing to stop discriminating based on protected class, despite that being illegal.

Quotas, affirmative action, and now DIE have all been ruled illegal bigotry by the US Supreme Court, but Democrats can’t and won’t stop being bigots.

I can’t take a party who has flouted civil rights laws my entire lifetime whining about “muh rule of law!” seriously.

This is a tangent, but your setup sounds interesting to me — would you share more about how to configure such for myself?

Yes — the Roman Empire was in a period of retreat and reformation when it formed.

That’s my point:

The US has neither entered an empire nor the decline post-empire because it gained territorial holdings over the past century — that’s consistent with late stage republics as well.

I agree with you right until the end:

This is closer to the end of the Roman Republic — and strange events such as threats against Greenland and Canada can be interpreted in that context as well.

That is, we’re not seeing the downfall of the US empire but the downfall of the US republic that will eventually retreat to North America and retrench its position — hence moving Canada and Greenland (and Cuba and…) from informal relations with a hegemon to formal vassals of an empire. And emerge as an actual empire, much like Rome before.

While I understand the emotional temptation, using the word “empire” in place of “hegemony” can lead to inaccurate predictions — eg, total US collapse versus pullback for retrenchment and reformation.

Robert Pape has been giving interviews on Mario Nawfal’s channel, among others.

I also appreciate his perspective.

https://www.youtube.com/@MarioNawfal/search?query=Robert+pap...

Yes — the US has helped Taiwan build them all over their mountains.

No — I meant worse health.

You die in both scenarios, you live slightly longer (eg, 8 months versus 6 months) and during the overlap, you are less healthy than you would have been. Losing 6 months of seeing family, enjoying trips, etc due to side-effects from treatment just to live 2 months extra, but incredibly ill the whole 8 months.

In my perspective: you experienced less total QALY by taking the treatment and had more time, in worse health. Even counting those 2 months at zero, the non-treatment option experiences more vitality, in total — because 2 sickly months at the end doesn’t offset 6 months of lost vitality immediately before.