From ChatGPT to Devana or how to go from "fast-think" to fighting fakenews from GPT4

More than 2 years ago. The startup Scriptor Artis wrote to OpenAI for the honor of using its revolutionary GPT 3 AI. Its idea? To create an intelligent system accompanying the researcher in his search for information until the written production. How? By allowing the live popularization of scientific and technical contents, while proposing qualified and relevant readings to increase the critical spirit of the researcher.

At this point, it is a question of writing a concrete file, presenting a major value: the ethics of the project.

At the end of 2022, ChatGPT explodes into the open, taking on the air of an information fast food: the new fast-think? It looks like it. Because the goal seems to be to give an instant solution, regardless of its verification... Not a day goes by without the press reporting outbursts concerning unbelievable information, threats to humans...
Scriptor Artis can't believe it. What is certain is that its AI Devana is of major interest in this paradigm shift.
So, can AI, in this case GPT 4, really be an information qualification assistant? Or should we be afraid of a fake-world where it will soon no longer be possible to distinguish the true from the false?
In this conference, Barbara Delacroix and Marvin Sant explain how GPT works (3 & 4) and how they tamed (and harnessed) it to make it a formidable tool for qualifying information, fighting against fakenews and increasing human analysis skills! That's all it takes...