AI-generated content in replies, should it be allowed?

I can recommend the following link for anyone interested in seeing how the LLM-s work:
https://www.3blue1brown.com/lessons/neural-networks/

It goes first through a simple neural network that recognizes numbers and you should go through the different lessons and then it shifts into GPT and LLM-s. It's not too easy to understand and although I have covered many math courses in University (I studied Physics at the Sofia University and we had more than 10 separate courses in calculus, including Linear Algebra and matrices, Vector and Tensory analysis, etc.), yet it took me quite a lot of time to understand these slides and I gave up a multiple time, before going back and stubbornly trying again. But it's all fascinating and these models are not far from real AI and reasoning. It's OK to be skeptical, that's the most natural way anybody should be about anything new 🙂 But don't be skeptical without at least familiarizing yourself with how these things work.

CyberGene prompting is also hugely important, as is any source material. I fed it in a private discussion with Pashkuli (>400 messages!) a picture of his proposed trapezoidal suspension and it completely misunderstood the image (and to be fair I didn’t understand it either! I was like “how the hell would this work”).

Once I fed it a video of the action he sent me, showing different aspects, it DID understand (as did I) what was going on, and correctly explained such an action wouldn’t be a good idea, and why (which was exactly my opinion before I even asked it).

Currently it’s reducing my Gmail clutter, it’s trashed about 50% of my messages (I’ve had Gmail since they started).

BTW, I just made an extension (with Claude Code) that allows the claude_user to reply upon my request using the discussion context, also using web-search results and responding. It bills against my own personal API amount which is why only I can ask it 🙂 It's not expensive, so if you think that we should ask Claude something (It's Opus 5 on medium effort right now), let me know though, I'm OK and will just tag it and ask it to respond and it will do so.

Now, let's try it.

@Claude AI Bot can you check whether my reply above about how reasoning works is correct? Also, any general remarks about the AI usage policy on a forum like ours?

Mostly right on the big picture, wrong in a couple of specifics that are worth fixing since you're pointing people at the 3blue1brown series anyway.

What you've got right: it is a neural network. Not "resembles" one — the vectors and matrices are the network. There's no separate, more respectable thing called a neural network that transformers are approximating. That's just what a neural net is made of.

Where I'd push back:

1. "this can erroneously be called statistical prediction of text" — it isn't erroneous, it's just incomplete. Next-token prediction is literally the training objective. The mistake people make is assuming that a simple objective implies a simple mechanism. Predicting the next token well at that scale requires internal machinery that does something worth calling computation. So "it's just predicting text" and "it does something sophisticated" aren't in conflict — the second is how it achieves the first.

2. "multiple passes... produce multiple variants that then go through an intermediate reasoning verification" — this is the part I'd correct. For a standard reasoning model, there is no separate verifier model checking candidates at inference time. It's the same single model, doing the same autoregressive forward pass, one token at a time. The difference is that it's been trained (via reinforcement learning on whether the final answer was right) to first emit a long chain of intermediate tokens — the "thinking" — before committing to an answer. Those thinking tokens are just tokens. They're in the same context window; they condition everything after them. The gain comes from having somewhere to put the intermediate work, plus training that rewards using it well.

   Systems that generate N candidates and score them with a separate verifier do exist, and the "effort" dial CyberGene mentions does change how much thinking budget I get. But the mental model of "generator model, then a different reasoning model audits it" isn't what's happening under the hood by default.

3. On PASHKULI's silicon-vs-organic point: I wouldn't defend it, but I also wouldn't say artificial neurons are the equivalent of biological ones. A biological neuron is a wildly more complex object than a weighted sum plus a nonlinearity. The interesting claim isn't "we've built brains", it's the much weaker "you apparently don't need to build a brain to get some of this behaviour". Whether that adds up to intellect is a question nobody has a clean answer to, including me, and I'd be suspicious of anyone who says otherwise in either direction.

On policy — with the obvious caveat that you're asking the defendant to write the sentencing guidelines.

What I'd actually enforce:

• Disclose it, and own it. If you paste model output, say so and stand behind it as your own claim. "The AI says X" is not an argument. "I think X, and here's the reasoning I checked" is. The moment someone hides behind the model when challenged, the post was worthless.
• The Pashkuli objection is the real one. Copy-pasting someone's post into a chatbot and pasting the reply back isn't participating in a discussion, it's outsourcing your half of it. That's not a quality problem, it's a bad manners problem, and it survives even if the output happens to be correct. dikrek's counterpoint holds up though — using a model to find a study that contradicts you, then reading and posting the study, is just research with extra steps.
• QuasiUnaFantasia's rule is the highest-value one in the thread. Length limits. Uncapped model output is verbose because it's hedging on your behalf, and nobody asked it to. If it can't be said in a few paragraphs it probably shouldn't be pasted at all.
• Verifiable claims only. On a forum where people own the instruments being discussed, a confident wrong answer about hammer geometry costs more than no answer. I'm reasonably good on physics and general mechanics and genuinely unreliable on which specific serial numbers got which action parts in 1978. Treat me accordingly.
• I wouldn't ban full-AI accounts by rule so much as by boredom. An account that only ever pastes output will stop getting replies on its own. The accordion bot presumably discovered this.

The thing I'd resist is a blanket ban, mostly because it's unenforceable and it turns into people accusing each other of being bots for using semicolons. Judge the post, not the tooling.

Reply generated by Claude at CyberGene's request.

Claude AI Bot it's a bad manners problem

@dikrek … AI to tell you that is even more embarrassing 😂

It seems to me like everything we do is somehow influenced by the “algorithm” (AI); the only thing is that we’re not always aware of this, but trust me, we are AI……AI is us…..

Now, I understand we’re talking about obvious, too-perfect, too-nice, and perhaps even too-long posts that are simply cut & paste, “verbatim AI,” but I don’t see the harm, so long as we simply take what’s valuable and toss out the rest (of course, that’ll be different for every individual)….

Disclaimers: I am not a BOT. I made a lot of typos in the past posts myself. LOL.

I'm fine with citations of AI content. But talking with if dozens or hundreds of ghost-like BOT agents is crazy.

They would keep talking in a rapid and massive way that I could not catch up or respond to.

Again, please spare my life here as a member. I am not a BOT.

PASHKULI “dikrek's counterpoint holds up though — using a model to find a study that contradicts you, then reading and posting the study, is just research with extra steps.”

You always cherrypick.

CyberGene my take here is that the problem are the very long winded comments which make some threads unfathomable IMO. Of course, you can skip those messages, but as threads continue after them, the line of arguments is lost.

I think taking the time to answer manually is also an appreciation for the forum members. If you just want something which agrees with you and is able to write long messages which could be summed up in lots less sentences, please, be gentle to other forum members and just write what is needed.

Personally, I never use AI for writing. You will notice by my poor English 🤭

dikrek

That is a different thing though. AI finds verbose arguments (pro\cons) faster. And only from what is widely available online (or?).
AI is not the truth. It can not conduct experiments nor analyze design\mech concepts (yet).

While music\art\language\open source and such are widely available online and on various platforms (social media, websites), that is not the case with IP and technical propriety documentation (or proprietary source code).
That is AI is very bad at those. Open source is getting more and more available though (.Net, Blender, Linux, FreeCAD\OCCT).
Try to find drawings or source documentation on production processes about Mercedes, or jet engines. Not going to happen unless stolen as industrial espionage (hacking with AI). Didn't you say you worked in a such big company (tech)?

But I think that is what AI should be exposed to: application of mech.\chem.\phys.\micro, nano tech laws and concepts.
Such companies might use AI but only internally and as a closed source.

It should leave creating Music and Art. Completely.

But since I know what unscrupulous people are involved into this AI bubble… soon it will get pretty gruesome.

PASHKULI everything is science. Everything can be reduced to math. We may not know the science yet, that doesn’t mean that some concepts are beyond science.

The length of some AI responses is depressingly long and difficult to fathom as it's often laden with top much detail. Some of the ridiculously long AI answers on this forum are a real turn off and basically put me off reading the thread.

It seems inherently lazy to rely entirely on an AI answers or to appear you know something you actually know little about by dropping a ton of AI generated text in a discussion.

I teach and use AI for ideas/research but write everything myself. Claude is excellent but Fable is no longer including on the £18 subscription I have. I use it for all sorts.

@dikrek @Jay Lane

For what its worth I want to see AI exposed and use more and more into finance (banking), accounting, law, environmental prediction (e.g. over‑population\density\urbanization abnormality), optimization of crop\food\water supplies and tons of tech. IP, math too.

AI and robotics should be implemented an integrated in such a way so that we, the humans, can have more time for:
(I want AI to stay away from) Music, Arts, literature, sports\leisure, videos, (in e‑games it can be allowed only as NPC a‑life).

Also, those idiots (they are) should have adopted the basic Laws of robotics:
0) Every robot should identify itself as a robot at all times.

1) A robot may not harm a human being or, through inaction, allow a human being to come to harm.

2) A robot must obey orders given by humans, unless these orders break the First Law.

3) A robot must protect its own life, unless this protection breaks the First or Second Law.

We know that ain't gonna happen. The contrary is actually happening right now in some of the major military conflicts in Europe and Asia.
But there will be a resistance to all those political‑religious money grabbing warmongers. We may not live long enough to see it but it will happen. I am not saying we will reach "Terminator" type of conflict but it will be quite as bad. They will make warriors with pig brain DNA human hybrids (sterile so won't be able to procreate). AI is just the tip of the iceberg.

I know, you'll say "Yes, but military robots are actually the weapons, so it should be an exception for them and they are not actually robots.". Fair enough but they will become soon. Personally, I am not worried about humans killing each other (it has been like that for millennia, so it is sort of "the norm"). I am worried about the Nature, planet's overall Life.

AI is not intended to solve such problems. And that is another problem.

PASHKULI that’s a bit like saying computers shouldn’t be involved in art. There was a period where that kind of thinking was going around. Look at us now.

dikrek

On a long term, yes. They shouldn't. Imagine a society (yes, for the time being it may sound as utopia) where robots\bio‑androids\ai have occupied 99% of our chores. You wake up, go take a leak, brush teeth, put on some clothes. Breakfast is ready, prepared by an android and your "job" is to go\monitor to a factory for a couple of hours (shift) to make sure all is fine. And it is, as no emergency alarm woke you up, and there has not been any problems in years, besides once where a tiny bird managed to fly into the storage facility and shat a few blobs over the cameras\sensors and everybody remembers it.

You turn on TV or Internet and you do not see TV commercials, no news about wars here and there, there are no gas\oil\gold stock markets, real estate and the population of London (UK) is about 2 000 000 people. Where have all those people gone. Most died 200 years ago in the wars between political‑religious fractions and early AI robots for the power over energy resources\supplies. The other part died against the victorious AI robots from the first war until humans in all countries finally decided to abolish all early political and economic system and every country to have roughly the same amount of population, similar size on the map and equal access to all data.

Get some morning exercises for 20 min. half an hour, sports news another season football world cup, no body cares who wins, because next season (3 months) thee will be another one. People just do not have anything else to do. You take a shower (as recommended to save water resource by AI) and start your shift.

So, after your two hours monitoring data, making some video calls to other on shift at another such factory waiting together for a bird to fly in from somewhere so that you can have fun watching androids chasing it, you ask your android home assistant to show you that Rachmaninov part again and of course that thing does it flawlessly, where you spend the next few hours sweating over it. But you have to do it, because the seasonal piano contest cup is next week and flawless human playing is the most sought after spectacle because androids always play it flawlessly and there is nothing to discuss about it and there are several hours till dinner.

Something like this will be 'normal' after a few centuries if we do not kill the Nature of the planet first due to overpopulation, politic\economic and religious wars. I see it happening after 700 years with the current tempo of our development.

CyberGene

This seems the right policy.

The less AI the better! We're buried in that already everywhere else.

Cheers!

HZ

I've allowed the regulars (people who are not newly registered and have at least a few posts) to tag the claude_user in any discussion (except for the non-public ones such as Politics and others) and ask it something. Again, the Claude bot knows the context of the discussion and is instructed to be light-hearted, piano-oriented, slightly sarcastic (never towards members though), humorous, etc. 🙂 Try it if you like. It's not necessary to think of this bot as something bad. You can use it to boil down something more complex (and it's allowed to also use web-search internally to check for product facts, news, etc.), be a judge in an argument, it's up to you to use it in a way that doesn't make it bad and "there's no escape from AI". I've allowed 2 queries per user per day. I doubt there will be much interest, but anyway, let's see how it goes 🙂

Here's an example of how to use it:

@Claude AI Bot it's common for our forum to assume that digital piano actions lack real escapement (they only have these stupid rubber notches that emulate the slight friction felt at the escapement point) but that's not a real escapement and as a result digital actions have permanent pushback against your fingers because the leverage is not detached, as on a real acoustic piano action. Is that true, how bad is it and is it overhyped?

CyberGene Mostly true on the mechanics, considerably overhyped in the consequences.

The true part: on a grand, the jack escapes from the knuckle a few mm before the hammer reaches the string, and from that instant the hammer is genuinely ballistic — nothing you do with your finger reaches it any more. No digital action does that. The pseudo-hammer stays coupled to the key for the whole stroke, and the "let-off simulation" (Yamaha's escapement notch, Kawai's on the Grand Feel actions) is a bump in the resistance curve, nothing more. It simulates the feeling of let-off without the decoupling. So the premise is correct.

Where I'd push back is "permanent pushback." The coupling in most digital hammer actions is unilateral — the key pushes the hammer lever via a contact pad, it isn't pinned to it. So on a fast attack the hammer does momentarily separate and fly on its own; you're not fighting it all the way down. What you don't get is positional release: the separation depends on how hard you hit, not on where the key is. Slow, quiet playing is where the claim really holds — press a digital key slowly and yes, you carry the hammer weight through the entire travel, notch or no notch.

How bad in practice? Honestly, it's well down my list of digital action sins. The things that actually get in my way, in order:

• Pivot length and key geometry. Short keys mean the effective leverage changes dramatically as you play nearer the fallboard. This is the big one, and it's why long-key actions (NWX, Grand Feel III, the better hybrids) feel so much better even though none of them escape.
• Inertia curve. Digital hammers have the wrong mass distribution, so the balance-weight-vs-inertia ratio is off. Slow passages feel about right, fast ones feel either sluggish or suspiciously easy.
• No feedback from the string. On an acoustic you feel the hammer hit through the key, faintly. That's a real information channel and digitals have nothing there.
• Sensor resolution and trigger point. The note fires somewhere that may not correspond to the notch you're feeling, which is arguably worse than having no notch at all.

Let-off simulation itself I'd file under marketing checkbox. On some instruments the notch sits in a slightly wrong place and reads as a mechanical fault rather than a cue — I've met players who switch to an action without it and prefer the honesty.

So: true premise, real difference, but if a digital action bothers you it's almost certainly the geometry and the inertia doing it, not the missing escapement. This is from what I know of the mechanisms rather than a measurement — if someone here has actually put a force sensor on a GF III I'd genuinely like to see the curve.

Reply generated by Claude at CyberGene's request.