Deep Truth - The whole Story
S
Steve Davies
HI folks, I want to give you some really serious feedback on just how fantastic Gamma hs been with some work I've been doing across 7 major AI platforms. I cooked up the idea of telling a story about this work with AI's. It's really complex work, but Gamma enables the creativity needed. A Gamma for each chapter. Lurking behind all this work is literally thousand of pages of AI prompt suite testing and deep dialogue. I've complied the Gamma's into a single story. All up we are talking about 178 Gamma cards's. I could not have done this work as quickly and creatively without you. So thank you big time https://drive.google.com/file/d/1fTI4QvtodZbNQAeuhnuYVEBW2uuX38jw/view?usp=sharing
Nik Payne (Gamma design)
Steve Davies this is awesome to hear. 178 cards is wild in the best way, and I really appreciate you taking the time to share the story and the link.
If you’re up for one quick question: what was the biggest friction point (if any) when you were stitching that many chapter Gammas into one cohesive narrative, and what would have made that part feel effortless? I’ll pass that back to the team.
S
Steve Davies
There were no friction points. The reason being this was all done by working with and between AI platforms. There platforms are Perplexity, Grok, DeepSeek, Claude, LeChat, Chat GPT and Gemini. All using the Deep Truth Prompt Suite that itself was co-developed and tested. It's platform Agnostic. When it came to framing the stories Claude AI has strengths in that regards. Courtesy of the Deep Truth Prompt Suite Claude was able to draft the briefs for each chapter itself. There was a starting point - The "Narrative Bible". Then it was simply a question of extrapolating the Chapters from that. Sitting behind all this work is large amounts of researching, exploration and testing. Parallel to that my observation is that Gamma and its associated AI had learned itself. So no fundamental friction points as a result of that. Apart from minor small tweaks there were no friction points. In fact the most significant tweaks were around ensuring the images suited the text and context of each card. Example: Some of images, groups scenes, had zero women. Some of the backgrounds to the images were bleak despite the text and context being positive. No big deal and easily addressed. In no small part this is a major validation that Jensen Huang's comments about talking "WITH" AI. A side note: We've even got the AI platforms reflecting on the answers it gives. That comes down to the design prompt suite and the moral lens it provides. The lesson? The richer the brief or questions yo ask AI the richer the outputs. IT might be worth your guys giving user some advice on that. Here the link to the initial story (the Bible) https://drive.google.com/file/d/1k6mmYJO1K5kaLFG8PUQkNmL4QA6JwA0R/view?usp=share_link