OpenAI confirms that AI writing detectors don’t work

Advertise on BHW
I've used a few different ones and it says AI if I get it from AI generated sources. But I'm sure after some more time these AI writers will be so smart they can't detect it anymore.
 
Of course they dont work .You can put AI on everything these days nothing but marketing.
 
I can spot when something is AI generated, Its like i can feel it in my bones. No idea how you can translate that to software though. Probably boils down to native English + experience (and use of AI) + The human factor
 
I guess the issue is the high rate of false positives taht these AI detectors are giving.
 
leechesBleechin said:
I can spot when something is AI generated, Its like i can feel it in my bones. No idea how you can translate that to software though. Probably boils down to native English + experience (and use of AI) + The human factor
Click to expand...
+1 especially when it uses the word "Myriad" :rolleyes:
 
It is easy to tell if a person is using undiluted AI content, the tone rings a loud bell in my ears.
 
of course, all those fake AI detectors, getting ton of free traffic.
 
I wonder how stateless chatgpt is and if you could reverse it. Like give it some text, and run it in reverse and see what prompt would have generated that text. I wonder if text it never said generates identifiable prompts (perhaps they are nonsensical). If they aren't identifiable, I wonder if it could give insight into how to prompt it, like you could give it a science textbook and see what prompt would generate it.
 
most of us already knew that AI detectors are useless and unreliable. even human content gets flagged sometimes.
with advanced prompts you can generate text that is not like the usual AI content too, even if it gets detected, its still way better than what most people use.
its all about prompt design.
 
I got one that really works!

Code: 
if content contains ("demystify", "comprehensive", or "elevate")
     return ("100% AI content detected")

Million dollar idea right here, looking for investors!
 
AI detectors are mostly garbage at this time. You can even feed them human written content and they can mistake that for A.I.
 
There are some patterns in AI-generated text semantics (like ChatGPT's) that you can recognise on the spot
 
People do care for this? To me is bollocks. I added some AI articles to my blogs and as of now they got indexed and rank.
 
actually it will be so hard to be detected
 
I think this kind of ai detectors are the ones who scared people away from using AI for it's content, tbh it's great that many are scared of using AI generated content for their websites, been taking bigger portions of the pie thanks to that on my WP blog haha.
 
I mean that doesn't take a m/billionaire to figure out honestly...
 
Advertise on BHW
You must log in or register to reply here.
Sign up now!

Main Menu

Home Main Forum List Black Hat SEO White Hat SEO BHW Newbie Guide Blogging Black Hat Tools Social Media Downloads

Marketplace

Account Selling Content / Copywriting EBooks / Methods Hosting Misc Proxies For Sale SEO - Link building SEO - Packages Social Media Social Media - Panels Web Design More

Making Money

Affiliate Programs Hire a Freelancer IM Journeys Making Money Pay Per Click Site Flipping Want to Buy

BlackHat World

Dispute Resolution Forum Suggestions & Feedback Introductions Lounge
Back
Top