Wow this text is horrible to read and has all the AI smells. I'm quite surprised because Anthropic as a company has enough self awareness and sophistication to avoid this, it affects their brand after all.
Here are the AI smells
> You don't set out to buy millions of tokens. You set out to build a feature, finish a migration, or run a task. The token count is whatever the model needed to get there.
You can see how forcefully punchy and grandeur it reads. Pangram also flags it as 100% AI written.2
I guess it shouldn't be surprising that Anthropic uses their own (really bad at writing) models to write their dev blog, but it sure is disappointing to read slop from a company that can easily afford human writers.
I don't know exactly what it is, but reading AI written text just makes me acutely nauseous, and it only seems to getting worse. It's not because I'm anti AI (I'm not), nor because I can tell the author didn't put in the effort to write it themselves. I think its more related to the actual language being so convoluted and artifical (duh) that it gives me a physical reaction. I wonder if its a combination of being forced to read so much slop text every day, combined with the uncanny valley like feeling I get from it.
Hard agree on all points. It's a little scary actually how uneasy some LLM communication can make me, almost like it's trying to induce a glitch in my brain. I don't think that intent actually exists in current models, but there is precedent for using language in unconventional ways to induce certain mental states, for example the (hopefully not actually real) "suicide pattern" from Neuro-linguistic programming [0]
I work on a carpentry simulator, I'm adding a "scan my slab" feature right now that uses the actual grain from your wood. Here's a real sentence from Opus 5.5 earlier today: "A board turned end for end shows it turned, and one turned a quarter turn shows it turned 90°" I had to stop and ask it what the hell it was trying to communicate. I'm sure we've all got examples like this, but it really is nauseating.
I am starting to figure out why AI prose is so hard to parse. AI writes English like it writes code. Which is not surprising given how important coding use case is for LLM vendors. Terse, formal style that lays out information in the order of logic, not in the order of understanding. The reader unpacks sentences the way he or she unpacks lines of code. LLMs are probably specifically trained to reuse code, and flowery Claudisms like "load-bearing" are just "function calls" into a standard semantic library.
Another trait of AI prose is ambiguity. It loves using words or phrases that can are grammatically valid but have two different meanings in two different contexts, and it is up to the reader to look ahead in the sentence and then reinterpret the whole thing. The very common offender is present tense verb vs plural noun (e.g. "tracks" meaning "agrees with" vs "tracks" meaning "plural of track"). This ambiguous English is so statistically prevailing in LLM writing that there has to be a technical reason for this phenomenon.
Reading such text - efficient for LLMs to write, hard for humans to read - is extra labor. And it is no wonder that reader is pissed when suddenly the reading experience becomes an uphill battle.
The last LLM which was fine to read is Sonnet 4.6, but while it writes well, it makes factual errors often.
> I am starting to figure out why AI prose is so hard to parse. AI writes English like it writes code.
I think sometimes it just writes nonsense. As in there are words there, they're all strung together in a grammatically correct way, and the words are "right", but the sentences just don't contain any actual information.
There have been sentences and paragraphs that I have read over and over, and I just cannot extract any meaning from them. It's very, very strange, and I think that's the uncanny valley of it. It's all there, in the sense of words and grammar, but there's nothing there, the sense of knowledge and meaning.
I think stuff like “tracks” comes from leagues of the absolutely most condescending technically competent repository in the history of words and letters: stack overflow forums.
And some of the ambiguity I think comes from liability training sets, designed with a preference for not making direct attributions to parties who could claim they were incorrectly characterized. The habit can be seen in like half of Claude messages, where its responses have multiple inert objects or concepts being described through active personalized verbs. Like “the build rejected an undeclared package” as opposed to saying “the build failed because I didn’t update the packages.json.”
It is strange seeing prose from these models becoming progressively less human over time. Even the code opus 5.5 generates is kind of digusting and extremely robotic in a way that no human would ever write.
I just assume the training data is AI generated hence why it's so uncanny valley. Even though platforms that provide data like mercor and micro1 are supposed to provide "organic" data that isn't output using LLMs I've heard some people I know on those platforms say they generate their work using ChatGPT.
Wow this text is horrible to read and has all the AI smells. I'm quite surprised because Anthropic as a company has enough self awareness and sophistication to avoid this, it affects their brand after all.
Here are the AI smells
> You don't set out to buy millions of tokens. You set out to build a feature, finish a migration, or run a task. The token count is whatever the model needed to get there.
You can see how forcefully punchy and grandeur it reads. Pangram also flags it as 100% AI written.2
I guess it shouldn't be surprising that Anthropic uses their own (really bad at writing) models to write their dev blog, but it sure is disappointing to read slop from a company that can easily afford human writers.
I don't know exactly what it is, but reading AI written text just makes me acutely nauseous, and it only seems to getting worse. It's not because I'm anti AI (I'm not), nor because I can tell the author didn't put in the effort to write it themselves. I think its more related to the actual language being so convoluted and artifical (duh) that it gives me a physical reaction. I wonder if its a combination of being forced to read so much slop text every day, combined with the uncanny valley like feeling I get from it.
Hard agree on all points. It's a little scary actually how uneasy some LLM communication can make me, almost like it's trying to induce a glitch in my brain. I don't think that intent actually exists in current models, but there is precedent for using language in unconventional ways to induce certain mental states, for example the (hopefully not actually real) "suicide pattern" from Neuro-linguistic programming [0]
I work on a carpentry simulator, I'm adding a "scan my slab" feature right now that uses the actual grain from your wood. Here's a real sentence from Opus 5.5 earlier today: "A board turned end for end shows it turned, and one turned a quarter turn shows it turned 90°" I had to stop and ask it what the hell it was trying to communicate. I'm sure we've all got examples like this, but it really is nauseating.
[0] (only link I could find that includes the supposed pattern, maybe don't read if you've ever had suicidal ideation though just to be on the safe side) https://frankreport.com/the-suicide-pattern-of-nlp-was-it-he...
I am starting to figure out why AI prose is so hard to parse. AI writes English like it writes code. Which is not surprising given how important coding use case is for LLM vendors. Terse, formal style that lays out information in the order of logic, not in the order of understanding. The reader unpacks sentences the way he or she unpacks lines of code. LLMs are probably specifically trained to reuse code, and flowery Claudisms like "load-bearing" are just "function calls" into a standard semantic library.
Another trait of AI prose is ambiguity. It loves using words or phrases that can are grammatically valid but have two different meanings in two different contexts, and it is up to the reader to look ahead in the sentence and then reinterpret the whole thing. The very common offender is present tense verb vs plural noun (e.g. "tracks" meaning "agrees with" vs "tracks" meaning "plural of track"). This ambiguous English is so statistically prevailing in LLM writing that there has to be a technical reason for this phenomenon.
Reading such text - efficient for LLMs to write, hard for humans to read - is extra labor. And it is no wonder that reader is pissed when suddenly the reading experience becomes an uphill battle.
The last LLM which was fine to read is Sonnet 4.6, but while it writes well, it makes factual errors often.
> I am starting to figure out why AI prose is so hard to parse. AI writes English like it writes code.
I think sometimes it just writes nonsense. As in there are words there, they're all strung together in a grammatically correct way, and the words are "right", but the sentences just don't contain any actual information.
There have been sentences and paragraphs that I have read over and over, and I just cannot extract any meaning from them. It's very, very strange, and I think that's the uncanny valley of it. It's all there, in the sense of words and grammar, but there's nothing there, the sense of knowledge and meaning.
I think stuff like “tracks” comes from leagues of the absolutely most condescending technically competent repository in the history of words and letters: stack overflow forums.
And some of the ambiguity I think comes from liability training sets, designed with a preference for not making direct attributions to parties who could claim they were incorrectly characterized. The habit can be seen in like half of Claude messages, where its responses have multiple inert objects or concepts being described through active personalized verbs. Like “the build rejected an undeclared package” as opposed to saying “the build failed because I didn’t update the packages.json.”
Yes, BTW the ambiguity of past participle vs past tense is another exhausting verbal tick. "The ticket described issue was resolved." Ugh
It is strange seeing prose from these models becoming progressively less human over time. Even the code opus 5.5 generates is kind of digusting and extremely robotic in a way that no human would ever write.
I just assume the training data is AI generated hence why it's so uncanny valley. Even though platforms that provide data like mercor and micro1 are supposed to provide "organic" data that isn't output using LLMs I've heard some people I know on those platforms say they generate their work using ChatGPT.