Subscribe
Sign in
Factiverse AI-Editor — Find factual mistakes in AI-generated content
Do you wonder if your AI-generated text is correct? Paste it in the Factiverse editor, and leave it up to our AI to find factual mistakes. We will also search Google and Bing in real-time to provide links to credible sources and save you hours of research.
85
Replies
@maria_amelie congratulations on this launch. Waited for it. Looking forward to publish it on easy save ai 👌
Drop me the info here https://easysaveai.com/submit-yo...
Well done 👍 I
@maria_amelie @wealthyshezzy1 Factiverse AI-Editor is the world's first tool designed to verify AI-generated content, helping users ensure the authenticity of claims made by AI assistants like ChatGPT or BARD.👏
With the exponential growth of online information, the ability to discern trustworthy content is more crucial than ever. 👍 Factiverse streamlines writing and content creation by providing quick access to credible sources, saving time🍦 on manual fact-checking.
@maria_amelie @wealthyshezzy1 @antonikozelski thank you so much for the nice and wise words! The exponential growth of online information is a challenge in itself! I read that every 14 days ChatGPT outputs more than what was published since the invention of the printing press in 1436..! And then add falsehoods and hallucinations to that..!
@maria_amelie
Congrats! Have upvoted and registered.
We all need your service. As a journalist myself, I have been looking for something like this.
I thought this can be used to fact-check existing content, but I guess it's for new material. Otherwise, when I ran one of my own newsletters — http://sreenet.substack.com — through Factiverse, it ended up citing my own newsletter instead of other sources. Tried the same with a WSJ essay and has a similar issue.
Looking forward to trying it with my next newsletter, BEFORE I publish!
Hey Factiverse team, congratulations on your launch!
The concept of an AI-powered fact-checking tool is indeed intriguing. I'm particularly interested which topics the AI Editor is especially good with if there are any?
@ivan_pasichnyk Hey Ivan and thanks for your feedback! It is currently performing best on well-covered topics such as climate change, news and popular science. The model is struggling a bit when there are numerical claims, so that is a topic we are putting more focus to in the coming months!
💡 Bright idea
Congrats on the launch! Cool project, verifying content is really hard. Is it possible to connect your own knowledge base of facts? To tune this to your own proprietary data.
@p1f Right now we add API from Google and Bing. But we can add other databases - would love to chat about that!
❓ The problem
Exponential growth in amount of information, LLMs are hallucinating, Fake news.. You know it
✅ Our solution
* Automatically detect controversial claims in any text
* Live search for credible sources - on both sides!
* Generated snippets for faster assessment of different sources.
👻 How it works
☝️ Access AI Editor here (https://editor.factiverse.ai/home)
📝 Paste in output from ChatGPT or write any text
✅ Click Analyze
🤖 Our AI spots controversial claims, finds you credible sources and provides a quick overview of the information out there
We’d love your feedback and thoughts!🙏
Very smart thing to have in light of chatGPT and other tools alike trying to just make stuff up sometimes (I already have been on the receiving end of it). Would totally use it
Congratulations! Very smart idea. Use AI to find AI mistakes. Would be interesting to know how that works behind the scenes and how your AI engine works!
@yuriy_yunikov We were careful to choose which data we can train on, mainly articles from highly credible newsrooms and fact-checkers. First we identify a fact-check worthy sentence, and then we search in Google and Bing to find relevant sources. We are not just showing the first thing that pops up there, but sort results based on the historical training data from newsrooms. So if historically certain websites are more credible on certain topics - we will show them in the results on that topic:)
@kevin_t_ Thanks, Kevin! You are absolutely right, we'll update with your suggested text!
@maiuran_loganathan1 Thanks Maiuran! I sometimes think that ChatGPT and generative content can pivot and bring a boost to high-end and credible content creation(!)
@vladimir_zivkovic Thanks a lot! Please let us know if you have any feedback!
Congrats on the launch! A really cool idea, ironic how AI is being used to verify AI :) How are you able to evaluate if a source is credible though? Do you just have list of preset credible sources (or endings such as .gov) you work from?
@deds3t Thanks Dennis! Technology is a two-edged sword ;)
We're not blacklisting or whitelisting any pages - but every time we do a search we also look at historical credibility for the source. So if you have a history of being credible around climate change, you'd rank higher for that topic, but not necessarily for inflation or economy. Our CTO @vinaysetty can also add to this!
@staticmaker1 Thanks a lot, David! We tried not to re-invent the wheel, but to keep i clean!
@staticmaker1 @gautek Thank you so much. Our website has undergone so many transformations, as well as the product - it is a never-ending process.
Congratulations on the launch of Factiverse AI-Editor! This looks like a great tool for improving AI-generated content accuracy. Looking forward to seeing it in action!
Factiverse AI-Editor