doraby

I got so bored writing marketing copy that I ended up creating a scenario for the landing page: how a random AI agent from 2125 discovers our app in the internet archives and try to use it.

Ever wondered what will happen to your apps a century from now?

I tried the first 3 paragraphs from your website and got "The classifier considers the text to be very unlikely AI-generated."

The next 3 paragraphs from "Our classifier has a number of important limitations..." showed "The classifier considers the text to be possibly AI-generated". This inconsistency does not look reliable, it would be great to see more details about the result.

Thanks you for pointing this out. Now these cases are post-moderated, but we will add the pre-moderation.

Well, we do not place all bets on prompt. We want to help people moving towards their career goal, see gaps in knowledge and career opportunities, and this core part depends on other things.

Haha, nice. Half of this is very close to reality. The second half is almost entirely a hallucination :)

Great, this is our goal. TBH we are not there for the time being, but we strive to get there.

The first is pretty close to prompt. The second part related to a career seems to be a hallucination.

Yes, so far. We are working on other ways.

We’ve created GPT-3 based AI mentors that help people to find the best possible career and learn new skills faster than on Google or YouTube.

With each question in chat, AI topics suggestions and questions become more personalized and generate insights on topics a person delves into. Each question updates skills makes career prediction more accurate, and shows blind spots in skills that need to be built up.

Remarkable idea! I like the "Ask your parents" feature.

Interesting thought, thanks for sharing. But I disagree, it doesn't look like that.

I can confirm that we are not Russian and use the Google firebase authentication service with default settings, like Canva, Figma, Miro and any other online tool on the Internet.

Yes, absolutely. I just realized that removing this information would even be beneficial. Actually, I'm removing it right now :)

I see. In this case, it only applies to the profile picture or avatar.

Actually, this is a quite reasonable remark about the name, we should have elaborated on this part. We will change all names to anonymous.

We've deleted all information about you. Thanks for giving it a try :)

We will definitely add this possibility, thanks for the feedback

Send us a request with your email to info@unschooler.me. Or just send a message to the chat on any page.

Yes, we use standard firebase auth service and display the avatar and profile name. This is my profile https://unschooler.me/profile/lena-dorogenskaya-1578. Also, under the hood we are analyzing tags of videos that users liked. Without this, the prediction is impossible. Actually, I don’t know how photos appeared in the scope, it must be a mistake or default settings. We will fix it.

While making learning platform, we noticed the four types of lecturers:

1. Boring with humor-free monologues. 2. Interesting lecturers talking to their audience like David J. Malan. 3. Exciting lecturers started from a question like Robert Sapolsky. They repeatedly returned to the audience examples, making a live lecture more engaging than a recorded webinar. 4. Richard Feinman. Just an incredible level of humor and clarity.

Considering the longread or youtube video as a lecture form, there is no opportunity to engage readers besides comments, but people mostly do not tend to comment. Honestly, comments on youtube are not for conversation, but for wit contests.

On the contrary, tests and quizzes force active thinking because the purpose is to answer. But there is a lack of information.

We crossbred longread with tests and Feinman by making the publishing platform with interactive tasks. Each task combines questions, answers, challenges with a content. And everyone will have the opportunity to learn something by explaining according to the Feynman learning method.

What do you think of this approach?