UnAIify's profile picture. Un•AI•ify helps you spot AI-generated text and reduce heavy rhetoric from online content. Improve writing skills and bolster know-how about persuasive tactics.

Un•AI•ify

@UnAIify

Un•AI•ify helps you spot AI-generated text and reduce heavy rhetoric from online content. Improve writing skills and bolster know-how about persuasive tactics.

고정된 트윗

Except it's all the LLMs.

UnAIify's tweet image. Except it's all the LLMs.

Un•AI•ify 님이 재게시함

Online, any attention regardless of whether it's positive or negative is good. Raging against someone's take so you repost it with a snarky response? Good. Or if some poaster's hot take you love and you like and repost ... good. If it engages one group and enrages another, you…


Un•AI•ify 님이 재게시함

I think people are still unprepared for a world where you cannot trust any v̴i̴d̴e̴o̴ content, despite years of warning. FTFY. cc @UnAIify


Un•AI•ify 님이 재게시함

I think people are still unprepared for a world where you cannot trust any video content, despite years of warning. Even when Google & OpenAI include watermarks, those can be easily removed, and open weights AI video models without guardrails are coming.. 404media.co/sora-2-waterma…


This but for AI generated content in general. AI could end the Internet.

Sora 2 is for humans what email was to direct mail. An infinite amount of garbage video will cause virtual ennui and drive people back to experiences on their doorstep. It's so perfect. And I can't wait for the revolt. @rorysutherland



Once you recognize the "It's not X" pattern, you start seeing it everywhere (and more and more every day). The slopocalypse is nigh.

UnAIify's tweet image. Once you recognize the "It's not X" pattern, you start seeing it everywhere (and more and more every day).

The slopocalypse is nigh.

New existential crisis just dropped: A social media-born video slopocalypse: AGI born from a Sora 2-fueled, reinforcement learning loop, trained on TikTok-style AI short-form videos. Good god.

UnAIify's tweet image. New existential crisis just dropped:

A social media-born video slopocalypse: AGI born from a Sora 2-fueled, reinforcement learning loop, trained on TikTok-style AI short-form videos.

Good god.

yes because engagement in a tiktok style feed is the greatest RLHF mechanism that has ever existed. the models will improve shockingly fast with this feedback loop



This is the visible "value" of products like Sora 2 ... trifles, memes, ephemera. OpenAI's valuation is a half-trillion. x.com/venturetwins/s…

My favorite new trend in the Sora app is putting Pikachu in every movie. This is “Saving Private Pikachu” 👇



Un•AI•ify 님이 재게시함

This is the unrecognized (so far) problem with AI writing. AI-generated writing takes a shortcut to avoid the pain the writer needs to go through to clarify their thinking. Worse, AI writing tends to adopt the muddy, heavy rhetoric of popular, people-made writing (the stuff it…

justinowings's tweet image. This is the unrecognized (so far) problem with AI writing.

AI-generated writing takes a shortcut to avoid the pain the writer needs to go through to clarify their thinking.

Worse, AI writing tends to adopt the muddy, heavy rhetoric of popular, people-made writing (the stuff it…

Un•AI•ify 님이 재게시함

The strangest thing about seeing model outputs everywhere is wondering how many people even notice. Maybe ten percent? Language is not just a carrier of meaning, it is a scaffold for many things. This is a live experiment, so we will discover the second order effects together.

Starbucks is closing two of the locations we have leased to them. They didn't bother to notify us. So, that's fun.

MattPaulsonSD's tweet image. Starbucks is closing two of the locations we have leased to them.

They didn't bother to notify us.
So, that's fun.


But inquiring minds want to know if _this_ like from Andrew was an intentional nod to AI writing or ... "Language is not just a carrier of meaning, it is a scaffold for many things."

The strangest thing about seeing model outputs everywhere is wondering how many people even notice. Maybe ten percent? Language is not just a carrier of meaning, it is a scaffold for many things. This is a live experiment, so we will discover the second order effects together.



The whole "AI detectors don't work" discussion is moot. All that matters is whether or not people become prejudiced against some content because they think it's AI. (This is a major use for tools like unaiify.com, to lower the probability)

A note about AI writing checkers: Nothing will ever come back with a 0% chance of being AI. That's because AI is trained off of the writings of others. Typically anything under 30% is a confident no, and under 10% is not worth questioning.



Un•AI•ify 님이 재게시함

Made by AI? You don't say: "On the path to AGI, the gains aren’t just about productivity — they’re about creating new possibilities, sparking creativity, and bringing joy." It's not X. It's Y ✅ Triad ✅ Assumed an em dash. But could have been a semicolon. Such amazing…

UnAIify's tweet image. Made by AI? You don't say:

"On the path to AGI, the gains aren’t just about productivity — they’re about creating new possibilities, sparking creativity, and bringing joy."

It's not X. It's Y ✅
Triad ✅

Assumed an em dash. But could have been a semicolon.

Such amazing…

The fly in the AI ointment is "Workslop" (screencap from Morning Brew below). People using AI to create work artifacts that are half-baked, full of errors, passed along as "good" ... This "workslop" must be dealt with and managed by those who know better.

UnAIify's tweet image. The fly in the AI ointment is "Workslop" (screencap from Morning Brew below).

People using AI to create work artifacts that are half-baked, full of errors, passed along as "good" ... 

This "workslop" must be dealt with and managed by those who know better.

Because AI makes a probabilistic guess on word count.

why does all AI writing feel like a student trying to satisfy a word count?



Un•AI•ify 님이 재게시함

Things AI has ruined: 1) entry level college grad jobs 2) white collar back office job security 3) the em dash 4) “it’s not X, it’s Y” 5) startup seed round expectations


This is why people get up in arms about em dashes. "This is written by AI" is a symptom of a much bigger problem: Not knowing what to trust.

AI controlled Bot Farm. 100s of them tossing 1000s of comments and postings designed to agitate you. Polarize you. Offend you. Disempowering you.



"it's not X" has become a mindset.


Your AI is showing.

UnAIify's tweet image. Your AI is showing.
UnAIify's tweet image. Your AI is showing.

United States 트렌드

Loading...

Something went wrong.


Something went wrong.