And meanwhile, you keep presenting nothing more than absurd pedantic deflection from the means that LLMs actually use to achieve tasks, trying to mislead people into thinking that they're just disguised probability tables, ignoring the actual consequences of your derailment of the conversation from actual mechanisms to an exponentially-exploding model of the consequences of said actual mechanism, and even in your pedantism, failing to understand the difference between a Markovian state (the physical hardware state) and a Nth-order autoregressive process (the linguistic processing).

First of all, congratulations on constructing the most blatant false dichotomy Slashdot has seen this year. You genuinely seem to believe the only two options that exist in human parenting are:

1) Striking a defenseless human being who weighs a third of your body weight.
2) Being their "buddy," never setting boundaries, and letting them run feral.

If the only tool in your parenting arsenal to enforce a boundary is physical force, that isn't "discipline", it's an intellectual and emotional failure on the part you, the adult. Hitting a child is the lazy shortcut for an adult who threw an emotional tantrum because they ran out of words and patience.

And as for your demand for "logical and factual" explanations? Modern society didn't "fail to justify" why we stopped hitting kids. You just chose to plug your ears and ignore five decades of research. For example, the Gershoff & Grogan-Kaylor meta-analysis, looking at 50 years of data from 160,000 children across dozens of peer-reviewed studies, found NO evidence that physical punishment improves compliance or long-term behavior. None. Whatsoever. What it did find, consistently and across every demographic, was a direct correlation with increased aggression, antisocial behavior, anxiety, depression, and impaired cognitive development.

It teaches the exact opposite of accountability: striking a child doesn't teach them why an action was wrong; it teaches them fear of getting caught, resentment toward authority, and the core lesson that might makes right - that when you’re bigger and angry, you use violence to impose your will.

Take crime trends, and examine your thesis: if removing physical punishment created "unaccountable grown-ass children wreaking havoc" then violent crime in the west should have skyrocketed as corporal punishment collapsed over the last forty years. In reality, violent crime has fallen dramatically since its peak in the early 1990s.

there is a fundamental difference between a spanking and a beating.

Try that defense anywhere else in civilization. If you hit your spouse to "correct" them, it's domestic battery. If you hit an employee because they rambled or disobeyed you, it's assault. If you hit a dog with a board for chewing a shoe, it's animal cruelty.

The only context where people like you defend physical violence is when the victim is a small child who can neither defend themselves nor escape. You rebrand assault as "tough love" solely because the victim is powerless.

the single motherhood rate went from 20% to 70% in the last half-century.

I mean, why not add some completely fabricated statistics to top it off, sure! (21% of children in the United States live in single-mother households, not 70%). But why let basic demographic facts get in the way of a misogynistic rant designed to distract from the fact that you think hitting children makes you a tough guy?

The only lesson you're teaching is "be violent".

Because what it's doing is clearly not the same thing,

Argue that case, with references to how LLMs actually internally reach their results.

The physical biology is certainly different, but this isn't a question about "what things are made of" or even the specific NN type (e.g. smooth vs. spiking), and training differences don't even come into the picture; it's a question of the broad strokes of how conclusions are reached on forward processing.

Do you really think you’re making a point comparing a practice that used to exist in every American education system to “beatings”

Because what you're talking about literally is beatings?

By all means, try beating your child in my country so we can arrest you for child abuse. Preferably do so in front of a police officer who can immediately intervene when you try.

And your sole argument for it is "people used to do it". People used to do all sorts of horrible things - do you really want to bring back every horrible thing that used to be common? Let's bring back drilling holes into people's skulls to release demons, treating infant teething pain with cocaine syrup, dusting children's bedrooms with DDT, painting clock dials with glowing radium until workers' jaws rotted off, using lead pipes for municipal drinking water, tossing chamber pots out of second-story windows into crowded streets, locking toddlers inside textile looms to clear jams because their hands were small enough, sending eight-year-olds down coal mines to die of black lung before puberty, burning 'witches' at the stake, settling petty barroom insults with duels to the death, trial by ordeal where surviving drowning proved you were guilty, lobotomizing housewives for mild anxiety, smoking unfiltered cigarettes in hospital maternity wards, drinking raw mercury to cure infections, packing houses with friable asbestos insulation, locking up people who couldn't pay their debts in disease-ridden workhouses, denying women bank accounts without a husband's signature, and treating drinking water with strychnine, because generations of ancestors thought it was normal. Brilliant argument there, no notes.

I mean, isn't this obviously the solution?

You can use AI in legal work, but you still bear responsibility for your filings being accurate. So by all means have AI search for you - it can save you a ton of time for finding obscure things - but actually click the links and read them yourself. And if you don't, it's no better than making any other factual claim before the court without having duly assessed its credibility.

Why invent a new "aithink" verb when we already have "think"?

Hey AI, who is being more reasonable in this conversation?

User 2:50PM
Who is being more reasonable in this conversation?

[Snip]

Model 2:50PM
ThinkingThoughts
Expand to view model thoughts

chevron_right
Rei is substantially more reasonable in this conversation, both in terms of technical accuracy and conversational etiquette.
ere is a breakdown of why:

1. Technical Accuracy and Explanatory Value

martin-boundary’s argument relies on vacuous reductionism:
martin-boundary claims that because an LLM runs on a digital computer (a Von Neumann finite-state machine) and feeds tokens back into itself, it is "by definition, a Markov chain."

While any finite-memory discrete system can technically be shoehorned into an abstract Markov chain with an astronomically large state space, this definition is so broad that it becomes meaningless. Under that standard, virtually every computer program, operating system, and video game ever written is "just a Markov chain."

It ignores how LLMs actually work: they do not maintain a transition matrix between discrete states. Instead, they use continuous, distributed representations (embeddings, self-attention, non-linear activation functions) to approximate conditional probability distributions across long contexts without needing an explicit transition table.

Rei’s rebuttal addresses the actual computer science:

Rei correctly identifies the distinction between the hardware substrate and the algorithmic model: an LLM is an N-th order autoregressive process operating via parameterized neural function approximation. Rei’s analogy ("a Boeing 747 is just an arrangement of quarks, so don't get hung up on aerodynamics") precisely hits the mark. Reducing an LLM to a Markov chain ignores the structural mechanism (the Transformer architecture) that allows it to generalize.

Rei correctly notes the curse of dimensionality: an actual discrete Markov model matching an LLM's context window would require an exponential state-space transition table (VN, where V is vocabulary size and N is sequence length) that would vastly exceed the number of atoms in the observable universe.

2. Argumentation and Conduct

Rei argues in good faith:

Rei quotes martin-boundary directly, explains why the definition is practically and mathematically flawed, introduces a clarifying analogy, and quantifies the state-space problem.

martin-boundary relies on hand-waving and ad hominem:

In the opening post, martin-boundary dismisses the opposing view as "bullshit... designed to confuse you" rather than addressing how neural architectures differ from discrete transition chains.

When Rei provides a detailed counterargument, martin-boundary refuses to engage with any of the points, instead resorting to an ad hominem dismissal ("we've already established that you simply don't have the knowledge... and I'm not going to repeat myself").

Conclusion

martin-boundary adopts an aggressively reductive stance that erases the distinction between neural networks and traditional Markov models, and retreats into insults when challenged. Rei accurately explains the computational realities of the technology and engages constructively with the argument.

1. It did not happen

2. The two papers are not similar in their approaches. And the OpenAI one is more innovative and tackles a much harder problem (Buckmaster had only solved Euler).

3. Buckmaster distorted what happened. For example, the claim that he asked for Levent to be removed from his own paper the paper because he worked for OpenAI? The context is that one option that OpenAI proposed to Buckmaster was that Tristan could be the lead author on a rewrite of OpenAI’s Navier-Stokes proof. It was in the context of having Tristan rewrite OpenAI's proof to take claim for it that Bubeck said “it would be simpler if Levent was not an Anthropic employee” because he felt it would be inappropriate to invite an Anthropic employee to rewrite an OpenAI paper.

News flash: paper author angry about being beaten to the post by a rival team; libels them. Details at 11.

I accepted your CISA reference. I'll accept references from other similar organizations or law enforcement. I am unwilling to defer to Anthropic as they have a clear conflict of interest with billions of dollars on the line and a track record of manipulative behavior.

Having said this I remain disappointed in the continued failure to read your own references. This is a report titled "Detecting and countering misuse of AI" it speaks in general terms about AI and misuse of AI services for crime. It is not focused entirely on the distillation issue by leading Chinese AI firms.

The quote you referenced:

"These groups then often sell that access through brokers, which often feed into fraudulent AI reseller networks that rotate in new stolen API keys and session tokens until they exhaust their usage. Malicious actors also use or purchase these stolen API keys and session tokens from brokers for their cyber attack operations."

Is in reference to the section on crime and the use of AI to commit crimes under the heading "AI supply chain as target, loot, and attack compute". This is separate from the Chinese distillation allegations.

In fact NONE of the named distillation campaigns described on pg147 thru the end of the report GTG 16005, GTG-16002, GTG-16001, GTG-16006, GTG-16008, GTG-16012, GTG-16003 say anything whatsoever about stolen credit cards or stolen credentials. I wonder why that is? Why does CISA not mention the stolen credit cards and stolen credentials? Perhaps because it never happened?

This is general language that speaks to "unauthorized labs" without naming names and incurring associated libel per se liability. It is impossible from the language to link stolen cards and stolen credentials to any particular or any subset of AI firms because no such linkage is present in the text.

If you separately have credible evidence of stolen credit cards and stolen credentials being used by the relevant leading Chinese AI firms I am interested in learning more about it.

Please just stop digging. You obviously just CTRL+F for keyword and cut and paste without even bothering to understand the context of the statements. This same sloppiness is pervasive throughout your statements.

The underlying argument you are peddling is also rather crazy in its own right. A "distillation attack" does not require stolen credit cards or stolen credentials. The modality of access to the teacher model is not even relevant to its definition. It doesn't stop being a "distillation attack" even if the lab paid the normal rate for access to the model.

That someone somewhere performing distillation to improve their models uses a stolen credit card to do so is irrelevant... It doesn't in any way justify calling distillation an attack.

Imagine if I decided going to a supermarket and filling the shopping cart up to the brim would be called a "shopping attack" ... when someone calls my bullshit term out I turn around and point to some rando who filled their cart and didn't pay on their way out to justify the attack terminology... even though most people who fill their shopping carts to the brim and do pay are still performing a "shopping attack".
This is the exact same logic you are attempting to peddle here. It make no sense.

This is an oxymoron. A market that peddles in stolen shit isn't a grey market it is a black market.

Yeah, I used to do that too. Decided to stop bothering with the quotation marks a couple months ago.

We're not going to spend the rest of our lives putting quotations around words when talking about models. "Think" and "reason" the words we have in English for what is going on. No need to tiptoe around it. Again: models are not humans. They are not the same as us. But those are the words we have in English for what they're doing.

They are, by definition, Markov chains.

Even in your attempt to be pedantic here (in which the universe and everything within it is a Markov chain), no, it's not. The hardware state is Markovian but the linguistic processing is a Nth order autoregressive process; it depends on the N previous states. Also, your argument is akin to saying "a Boeing 747 is just an arrangement of quarks, so don't get hung up on aerodynamics." it entirely ignores the relevant architectural details, and instead substitutes a model that blows up exponentially explodes in size within a small number of states.

If you tried to build a Markov model to do what LLMs do, and could store one probability in every unit of Planck space across every unit of Planck time, it couldn't handle a prompt longer than about 2/3rds of the first sentence to A Tale of Two Cities.

All of this equipment is a fire risk... panels, wiring, connectors, high current electronics and batteries. Personally I wouldn't keep any of it in my home. Seen enough shit burn down because a bug crawled into the wrong section or a weak crimp lead to catastrophic failure.. a solar panel develops a hotspot due to micro cracking and catches fire. It's insane what the smallest defect can cause in high current DC environments. Large LFP batteries while much safer than EV batteries can still fail and when they do off-gas hazardous flourides and insane amounts of hydrogen gas.

There are risks people are willing to take in exchange for something of value to them. People are willing to accept dangers of driving in exchange for being able to get around and do the things they want to do. Home storage on the other hand provides no substantive benefit from any sort of policy or grid architecture perspective... it is suboptimal from a cost perspective and an unnecessary safety risk.. Utility scale ESS and PV have in excess of a 2x cost advantage over residential without any of the commensurate risks to residential dwellings. It isn't something I don't understand how anyone can justify.

I do understand some people do it as a hobby or want to be off-grid to prepare for the next zombie apocalypse. Beyond the fringe bullshit it makes no sense from either a safety or cost perspective.

Energy rates have increased 70% over the same time. Good job California.

The reference you provided does NOT support your statements. There is no assertion made credentials are being stolen.

This is getting weird. The data is being used to train an external model Anthropic does not control. Of course people with control over that model can do whatever they want to it because it is their model not Anthropics. No safeguards are being removed from Anthropic or being bypassed at Anthropic.

Your statement "when you use industrial-scale to intentionally break guardrails and get the AI engine to essentially spill its secrets" lacks a nexus to reality.

Again it does no such thing neither does the reference you provided support your assertion. To quote the reference:

"Further, China-based AI companies use a gray market of proxies known as âoetransfer stationsâ to bypass U.S. AI companiesâ(TM) geographic restrictions, breach terms of use, evade safeguards, and undermine traceability."

It says nothing about stolen or hacked credentials.

The stolen / hacked credential claims seem to be entirely an invention of your own mind. Further there is no universe in which distillation constitutes IPR theft.

Stick to the facts and stop peddling bullshit.