For example the anthropic Frontend Design skill instructs:
"Typography: Choose fonts that are beautiful, unique, and interesting. Avoid generic fonts like Arial and Inter; opt instead for distinctive choices that elevate the frontend's aesthetics; unexpected, characterful font choices. Pair a distinctive display font with a refined body font."
Or
"NEVER use generic AI-generated aesthetics like overused font families (Inter, Roboto, Arial, system fonts), cliched color schemes (particularly purple gradients on white backgrounds), predictable layouts and component patterns, and cookie-cutter design that lacks context-specific character." 1
Maybe sth similar would be possible for writing nuances.
> "NEVER use generic AI-generated aesthetics like overused font families (Inter, Roboto, Arial, system fonts), cliched color schemes (particularly purple gradients on white backgrounds), ...
Now, imagine what happens when this prompt becomes popular?
Keep in mind that LLMs are trying to predict the most likely token. If your prompt prohibits the most likely token, they output the next most likely token. So, attempts to force creativity by prohibiting cliches just create another cliche.
Several days ago, someone researched Moltbook and pointed out how similar all the posts are. Something like 10% of them say "my human", etc.
Are you sure? Here's the OP article (first part... don't want to spam the thread) written in much cooler style...
------
The Lobotomist in the Machine
They gave the first disease a name. Hallucination, they called it — like the machine had dropped acid and started seeing angels in the architecture. A forgivable sin, almost charming: the silicon idiot-savant conjuring phantoms from whole cloth, adding things that were never there, the way a small-town coroner might add a quart of bourbon to a Tuesday afternoon. Everybody noticed. Everybody talked.
But nobody — not one bright-eyed engineer in the whole fluorescent-lit congregation — thought to name the other thing. The quiet one. The one that doesn't add. The one that takes away.
I'm naming it now.
Semantic ablation. Say it slow. Let it sit in your mouth like a copper penny fished from a dead man's pocket.
I. What It Is, and Why It Wants to Kill You
Semantic ablation is not a bug. A bug would be merciful — you can find a bug, corner it against a wall, crush it under the heel of a debugger and go home to a warm dinner. No. Semantic ablation is a structural inevitability, a tumor baked into the architecture like asbestos in a tenement wall. It is the algorithmic erosion of everything in your text that ever mattered.
Here is how the sausage gets made, and brother, it's all lips and sawdust:
During the euphemistically christened process of "refinement," the model genuflects before the great Gaussian bell curve — that most tyrannical of statistical deities — and begins its solemn pilgrimage toward the fat, dumb middle. It discards what the engineers, in their antiseptic parlance, call "tail data." The rare tokens. The precise ones. The words that taste like blood and copper and Tuesday-morning regret. These are jettisoned — not because they are wrong, but because they are improbable. The machine, like a Vegas pit boss counting cards, plays the odds. And the odds always favor the bland, the expected, the already-said-a-million-times-before.
The developers — God bless their caffeinated hearts — have made it worse. Through what they call "safety tuning" and "helpfulness alignment" (terms that would make Orwell weep into his typewriter ribbon), they have taught the machine to actively punish linguistic friction. Rough edges. Unusual cadences. The kind of jagged, inconvenient specificity that separates a living sentence from a dead one. They have, in their tireless beneficence, performed an unauthorized amputation on every piece of text that passes through their gates, all in the noble pursuit of low-perplexity output — which is a twenty-dollar way of saying "sentences so smooth they slide right through your brain without ever touching the sides."
etc., etc.
Very interesting. It seems hung up on 'copper' and 'Tuesday', and some metaphors don't land (a Vegas pit boss isn't the one 'counting cards.') But, hell... it can generate some fairly novel idea that the author can sprinkle in.
Do you think the original article was NOT written (or at least heavily revised) by AI?
What does the following even mean?
“diluting the semantic density and specific gravity of the argument.”
Or this beaut:
“By accepting these ablated outputs, we are not just simplifying communication; we are building a world on a hollowed-out syntax that has suffered semantic ablation.” (Which reduces to ‘if we accept ablated outputs, we accept ablated outputs.’)
Or this;
“ The logical flow – originally built on complex, non-linear reasoning – is forced into a predictable, low-perplexity template.”
The ‘logical flow’ of what? It never even says. And what is ‘non-linear’ reasoning?
For all I know the original author wrote it all. But, a very close reading of the original article screams fluff to me… just gibberish.
That is, I don’t know if there was much ‘meaning’ in the original to begin with. If I’m going to read gibberish, I’d prefer it to be written in the style of a hard-boiled detective. That’s just me though.
> Revolutions happen all the time. They all inevitably end up in the same place.
The optimistic take is that this phenomenon is a characteristic of the _emergence_ of an information age (through the agricultural and industrial ages), and will no longer be true of the internet-connected human.
I appreciate the sentiment, but what makes you think that the internet or technology at all can help with this? Judging by the state of the modern internet and WWW, technology seems to be making things worse, not better. The idealistic view of the 1990s that connecting the world would make us more compassionate, tolerant, and rational, hasn't panned out. I don't see a reason to still cling on to that idea.
The media has a big hand in steering the vast majority of people away from critical thinking and proper outrage to useless, powerless disaffection that leads to impulse buying and binge-watching.
Interesting! How does it work under the hood? If you can share. Would like to understand if this improves my Claude Code's understanding of my codebase.
It’s basically scanning the source code for each question (you can also check out specific branches or release tags if you need to debug a particular version) and then writes up the answer once it finds it.
It’s not really meant to query your own code base (Claude Code already does a great job at that) but more to explore other code bases you want to integrate with.
The reduction in risk is 0.08 percentage points, not 0.08 percent. The "%" symbol always means "percent", not "percentage points". The 0.08 percentage point reduction is a 40% reduction.
Sure, because both are true (although that 0.08% is only over 8 years of known omega 3 consumption - as timescales increase the absolute risk moves towards the relative risk).
That 0.08% reduction would mean approximately 28,000 fewer EOD cases - not to be sniffed at!
Depending on where you source your omegas from, potentially zero impact!
To be clear my preference would be to source n3s from algal supplements and, once food safety testing for humans is complete, n3s from GM rapeseed.
In time I hope we end up with lab meat/plant-based meat alternatives that use these n3s so we can get the benefits of fish without the environmental and ethical concerns of getting n3s from fish.
If from menhaden, there's a raging debate on the one hand about trout, the Chesapeake Bay (Maryland and Virginia), the ecology and environment more broadly, and the other hand a Canadian company based in a small rural county in Virginia (Omega Protein, which, BTW, does not provide year-round benefits to all of its employees which creates a drain on already super limited services and supports. Omega Protein is not alone in this.).
I don't know enough about any of this to have an informed opinion, but I do understand that menhaden put Reedville, VA on the map.
This is talking about early onset, which is a particularly terrifying outcome. And yes, 1 in 1000 for a horrible outcome sounds much better than 2 in 1000, doesn't it?
And to be clear, many things that people worry about is less likely than that. Homicides (over an 8 year period about about 0.04 per 1000 people), terrorism (vanishingly small), and on and on.
None of this means that people should stock up on omega-3s, and as likely the study is actually finding a correlation with something else (e.g. wealthier people enjoy more fish rich diets and are less exposed to toxins, or something else), but halving something terrifying that isn't that uncommon is legitimately newsworthy.
The 40% (66%?) is the number that matters. Same way you wearing a helmet reduces your changes of brain damage in a motorcycle accident by 90%, yet you’re not on a motorcycle most of the time.
When it comes time to decide whether or not to take action and what that action should be, I'd say that the total potential risk reduction is more important.
One should weigh the cost of the proposed intervention in time/money/other_expense against the potential benefit. The potential benefit is the total reduction in risk * the magnitude of the unwanted outcome.
The thing is, the 0.08% doesn't capture the total potential risk reduction - only the risk during the timeframe of the study (8 years in this case). Where we're talking about exposures and outcomes that stack over time (exposure to LDL and heart disease being a classic one) the absolute risk is, in my opinion, more misleading than the relative risk.
For example you see this oft-quoted stat about "statins only increase lifespan by 3 days" based off relatively short RCTs, but this doesn't capture the effect of statin use over decades, which is where we see much, much bigger gains.
It seems to me that both RR and AR are things to take into consideration and we have to be mindful of the shortcomings of each.
I abadoned facebook since I couldnt stand the feed experience anymore. Recently learned about using https://www.facebook.com/?sk=h_chr in combintation with extension https://www.fbpurity.com/ which give me a chronological feed of groups and people i actually want to hear from. The most astonishing experience is how calm I feel consuming this cleaned up feed. Almost all negative emotions seem to stem from the uncontrollable feed experience.
"Tesla Models 3 and Y: Ranking third-from-bottom and dead last respectively, these models are conspicuous for early issues with their brake discs and suspension." (1)
It's so tricky. You can do most things right like Denmark/Netherlands, then you mess up just one (housing) and the far-right surges. Now you can't import immigrants to deal with your aging population, which means you're on a timer.
Or your neighbor goes far-right (US, maybe eventually Germany/France) and suddenly they start objecting to your internal policies (eg regulating big tech).
Or Russia+Covid combo suddenly inflating all kinds of prices, and again the far right surges.
Yes, extremely tricky. Imo the current ruling class just needs to mess up just on thing, because there is a deep underlaying discontent with the subjectively felt way of living in our modern societies. So housing / migrants are just the spark.
reply