DEV Community: Ben

What's going on with IT ?

Ben — Wed, 21 Oct 2026 07:53:51 +0000

It's like the red VS blue pill. The biggest taboo ever, the big cow in the hallway that no body wants to see.

The human. many many effort have been made to make good sharp and clean minds, they graduate from uni, parents give all they can for them... then what ?
because there is no proxy, no norm, no defend tools, no prevention, no nothing in between,
the business man abuses those little minds... whip them with AI along the way,

I changed 3 groups (each less than a year) of employers, because I don't accept from a business to come and handle my mind anymore. It's a viable asset that I will accept no power on this earth to come and play with. they are like yeaaaaa it's just python, it's just a tool, I also do python, I know loops.

For you naive thinkers out there, that never got the right vaccines (they never will probably because the AI lie is heading us straight to the wall), every group of people in the civilized world has a brotherhood like system, I'm not talking about some secret twisted group in the dark, I'm just talking about the natural evolution of things, doctors (medicine) had this from centuries now, lawyers, engineers ....

You ? open source, You work for free, you know what ? those little lines of code are the very seed of everything, they will never tell you that, why ? because no body keeps track of that... When Jhon Ali at 09-09-2010-9pm goes to Github and pickup your algorithm for his corporate giga program.

Someone from Oracle, From Airbus, From Sony will just pick up your algorithm for free. Now its called AI.

You know what's the heaviest hole in the equation ? governments. let's not talk about that.

WAKE UP.

Why we need an independent party to measure article readability

Ben — Tue, 20 Oct 2026 09:52:53 +0000

In the age of AI; the "dead internet theory" is REAL, it is no longer a theory;

For the years to come, people would be still relying on serious content from here and there.

The problem we need to tackle here and now, and with all courage is that a lot (most ?) of authors are losing their times writing valuable articles if they don't have a space in New York Times or so... Same goes on the other hand, these articles are NOT READ at all, probably by bots ?

This dramatic loss of human efforts should not go to un-tackled.

An independent organism that measure article readability (I don't know how, assuming the website insights are reliably, go checking proxies like Cloudflare ? ask for a transparent Google analytics dashboard from the site provider ?, etc etc this is the second problem for people who are willing to tackle this issue).

For the site provider to keep authors writing while these articles are under water, You say, it's their responsibility, well probably, but in many cases, authors do not know they are having no traction... while still hanging on by a thread.

ChatGPT text pre-analysis example

Ben — Fri, 25 Oct 2024 09:37:35 +0000

Introduction

To illustrate a useful example of Safe-Text, let's imagine you are building the next AI product. You are not an end user of the new impressive ChatGPT, but you are directing user questions somehow to ChatGPT. To value your work, you are providing your users with the possibility of sharing their Q/As so at some point of requests, you are caching their questions. KaBoom! you are so subjected to a multitude of vulnerabilities! Yes, you always should think of user inputs pessimistically.

Script kiddies are not rare, bad doers also can take advantage of any solution of this kind, they adore when you store user textual data. Even with innocent inputs, ChatGPT would perform very poorly when text is malformed, this can happen because of many many reasons, outdated browser, human mistakes, mixing languages etc.

Arbitrary text can be dangerous

You need to escape bad HTML tags, fix mojibake and Unicode, detect text language, remove zalgos, fix spelling and grammar etc.

This is hard to put in one solution, many open source solutions are offered, some of them are implemented in Ruby or Python, some in C or Rust and others might be in JavaScript and WebAssembly. Furthermore, many text processing tasks rely heavily on machine learning models, like language detection or correct grammar.

Safe-text API

We have put dozens of technologies and machine learning models in one single API ! All you have to do is calling:

/api/clean_text?text=hello%20world&models=FixMojibak,Punctuat

It cannot be easier, only two inputs, your text and a set of available models ! Despite the heaving back-end processing and the amount of models applied (You can apply most or all of them in one single call!) latency is so low compared to other APIs based on heavy AI models.

We also promise more models and better accuracy in the near future :)

Enjoy the free-tier today and have a more robust and secure digital product :)

https://rapidapi.com/bacloud14/api/safe-text/
https://safe-text-api.com/

Highest priest GPT

Ben — Fri, 11 Oct 2024 07:14:27 +0000

This is a social hack game for fun. Wow your friends by answering their questions from another room, making them think it's AI doing the talking – but it’s really you behind the magic!

repo: https://github.com/bacloud23/Highest-priest-GPT
website: https://wizard-gpt.pages.dev/

Introducing Safe-Text API

Ben — Fri, 19 May 2023 21:03:44 +0000

I made an API for text cleansing, most of Python enthusiast already dealt with text right ? Some among us have probably a web-app or other solution based on user content.

The thing about text, is that it is so fragile; It can break the whole website, or at least can have rest in DB as scrambled as the user posted it but finally ends again screwing UI and giving a bad end user experience

For this, there are already many solutions, some rely on complex #regex, basic or advanced algorithms, or even heavy machine learning models.
I wrapped dozens of solutions of a multitude of technologies, from #Python to #Rust to #WebAssembly all in one single API !

It is only accessible through #RapidAPI but it comes with a generous free tier 📷
Please, give us a start if you are already on
@rapid_api

If you feel like criticizing, please keep it constructive, I will be happy to follow with your ideas.

Current models:

FixHTML.
DOMPurify.
Linkify.
Fix mojibark.
Punctuate.
Decancer.
Remove bad words.
Remove sensitive data.
Strip tags.
Wrap text. (comming soon)
Detect Language.

Examples of effective cleansing

FixHTML
here is a para bold bold italic bold? normal?

turns into

here is a para bold bold italic bold? normal?

DOMPurify
very bad html (cannot post here)

turns into

safe html

Linkify
Any links to github.com here? If not, contact test@example.com

turns into

'Any links to <a href="https://github.com">github.com</a> here? If not, contact <a href="mailto:test@example.com">test@example.com</a>'

Fix mojibark

The Mona Lisa doesnÃƒÂ¢Ã¢â€šÂ¬Ã¢â€žÂ¢t have eyebrows.

turns into

The Mona Lisa doesn't have eyebrows.

Punctuate
hello world

turns into
Hello world.

Decancer
vＥⓡ𝔂 𝔽𝕌Ňℕｙ ţ乇𝕏𝓣

turns into

`very funny text`

Remove bad words

you are a bad ass

turns into
you are a bad ***

Remove sensitive data

My dads credit card number is: 5555555555554444

turns into
My dads credit card number is: ################

Strip tags

<a href="#"> this is my cat</a>

turns into
this is my cat

Wrap text

(comming soon)

Detect Language

Bonjour Paris

gives

french

Note that Safe-Text API is a gigantic wrapper on top of other bright solutions; We have glued many other bright libraries to work in one single API. It is a complex solution technically as many technologies (Python, Rust, Wasm, Node ...) are all working together in one single API call.

Link to doc: https://ns514514.ip-142-4-215.net/
Link to RapidAPI: https://rapidapi.com/bacloud14/api/safe-text

You can head to https://ns514514.ip-142-4-215.net/documentation to try the API.

Kindly