Posted by coloneltcb 1 day ago
That's not how it works today.
No sepulcator company gets profitable by shipping just a sepulcator. A sepulcator absolutely must have AI, monthly subscription, cloud services and - up until recently - has to be blockchain-based.
ADDED: Because it would make the writing friendlier to more people.
I agree about the authentic voice. At one point, I got pretty unhappy with a fairly junior editor's changes to something I wrote. I was just going to let it go but my manager took it up with the editor's manager in the vein of "this isn't acceptable."
Basically, a significant portion of the population doesn't like writing or isn't good at it and really wants a "get it done" button. I might not love it, but the market is there.
So Grammarly is addressing a very real need. Further, it's really the only way for them to stay relevant, because you're getting AI editing / writing features in Gmail, Docs, Office 365, etc.
To me it is exactly why this move doesn't make sense.
Why would I use Grammarly/Superhuman for writing with LLM assistance, when I have an out-of-box alternative that, at worst, is equal?
They can't even compete with pricing, because they need to use their competitor models
I think the answer is basically that they have brand recognition and they're trying to ride it. Right now, they have two bad choices: become irrelevant more quickly by having a product that's inferior to built-in LLM tools, or become irrelevant more slowly by having a tool that's comparable (and also works anywhere on the internet, not just on specific websites).
They are a feature, not a company, with my apologies to Jobs. To your point, software and tools with native writing functionality can incorporate their own LLM support, as can native apps on mobile and desktop. Anything local will eventually be on device imho as model efficiency improves, or perhaps in browser (if not making API calls).
I did write for a while for a tech site that had some Wordpress add-on that was oriented to making my writing, I guess, more friendly to an 8th grade level. I ignored it.
I don't know what it is, but would love to hear others' ideas.
Resumes written by LLMs and read by LLMs
PR summaries written by LLMs and read by LLMs
Emails written by LLMs and read by LLMs
...
Everything could just be a few bullet points... these things were already 90% posturing and trying to sound fancy by using convoluted sentences and big words, now that it's been automated what's the point
On top of that, Grammarly's 3 products (Coda, Go and Grammarly itself) all look the same. What are honestly the differences between these three:
* Product 1: Everyone’s favorite AI writing partner.
* Product 2: The all-in-one AI workspace for teams
* Product 3: AI that works in every app you use
And then the packaging done per Enterprise features and metered.
For instance, if you have a misspelled word, and the correction options come up, you can't get out of them and return to where you were by using the keyboard. You can hit Escape to close them, but it doesn't restore your place in the text field, so you have to use your mouse to get back where you were.
As a programmer who tries to use the keyboard as much as possible, this (incredibly easy to fix, I'm sure) bug drives me crazy! Almost enough to make me go back to Grammarly.
The quality is unmatched because it uses SOTA models like GPT-5, Claude, and Gemini.
Yes, you can use your own API key as well.
I know it has a positive connotation with super heroes in US culture but for me it sounds like Übermensch. Especially as it is the direct opposite of "subhuman".
Plus outside of tech bro circles, people either actively hate generative AI or are at least super annoyed by the over-hype of it. Duolingo went all in on AI and got a huge shitstorm.
Branding your company on a current hype that might either burst soon or/and leave lots of people unemployed is maybe not a wise decisions.
I'm not sure about this. I'm a US citizen, but it absolutely does not have positive connotations to me at all. It has very negative ones.
Yes, I am. Born and raised in the US.
There are instances where the term is used in a positive sense, yes, but those are limited in scope. "Superhuman strength" rather than just "superhuman".
"Superhuman" on its own is a term that has long been tightly associated with a wide variety of horrible things. Eugenics, for example.
GP doesn’t imply Nazis used ‘Supermensch’, just that the ‘superhuman’ translates to übermensch and that the branding might evoke this concept for European ears.
It failed for financial reasons and the rather harsh environment. They ditched the vegetarianism and started selling meat to get some money, spiraling into alcohol and morphine abuse. In 1889 Bernhard killed himself with strychnine, after which Elisabeth started her career as a fake chronicler by writing a book aimed at creating a much nicer and 'Aryan' image of Bernhard and the colony than the truth would have allowed.
As you allude to, Friedrich Nietzsche poured buckets and buckets of abuse over people like his sister and other german nationalists, refusing for the entirety of his life to identify as german, and towards the end of his life he even claimed to be a polish nobleman, free of the tainted blood of the germans.
This is sort of a reverse version of the very common trend of American political correctness / sensitivity language being exported around the world. Our ancestors committed heinous crimes, therefore we get to tell you how to speak, even though you had nothing to do with it.
Person above argues that the words are different therefore such connection can't be made which is just... wrong because they reply in a thread where someone literally said they made that connection.
In short, we're explicitly talking about what Europeans see (me too, I'm not German), not what Americans should do.
The comment I'm replying to says, verbatim, "hey maybe this specific word shouldn't be used" (as a paraphrase of that commenter's understanding of the argument being made by the German). That is what I'm responding to.
If someone says a particular word or phrase is problematic for them, no one can tell them they're wrong. You cannot dictate how other people respond to language, and it's really weird to see people trying to do that.
What I object to is the implication that Americans should punish themselves by refraining from using normal words in their own language because Germans feel bad about something Germans did.
Shame for what Germany did during the Nazi regime is something for Germans to bear, not Americans. We are not at fault for that, and we have no obligation to change our own culture to accommodate your guilt.
I am not sure how important the German or general European market is so hard to say whether it even should be a consideration for Grammarly.
That said the ideas of some people being intrinsically better than other people isn't specific to Germany. Eugenics used to be popular in many countries including the US. It is very advisable for other countries to learn from German history so our mistakes are not repeated.
I'm sure a large population of second language English speakers is a huge market for grammarly, no?
Superhuman's entire value proposition was "staying out of my way".
All I can do is hope that the latter rubs off on the former, and notthe otehr way around.
Now I'm considering vibe coding a replacement that just does grammar fixing...
Ironic, because I don't want the AI writing, only the grammar checking
Search results are optimized based on inferred intent, and the intent of most people searching for "superhuman" will be the Grammarly app.
You can foresee a challenging future for the Grammarly product for a long time. Now that the "improve writing with AI" feature is everywhere, there are fewer reasons to pay for their subscription (e.g., I didn't renew this year because I have multiple AI subscriptions, and Grammarly was the least critical of them).
However, for me, the main advantage of Grammarly was the user experience of having mistakes and suggestions inline and just a click away while editing, as well as the quality of the suggestions (with an LLM chat, there's a lot of trial and error and junk you need to filter out).
I understand their move, but I wish they had developed a good minimalist native text editor with the same Grammarly suggestions and click-to-correct interface.