On Saturday, June 26, 2021, at 9:10 PM, John Rose wrote: > Splooging text to images and vice versa is not close to AGI it's merely a > fancy type of protocol translation. You keep saying DALL-E DALL-E DALL-E what > the f. else does it do?
..... EVERYTHING DALL-E and UE4 are 2 of the most advanced algorithms I've ever seen. DALL-E can be asked for an elongated elephant with 9 legs with pikachu riding it walking into a mouth of a giant frog that is in black and white. It can generate the rest of an unseen images using similar experiences. It is making discoveries when it can know the rest of the prompt without having seen the actual rest of the prompt. You can feed it any question/image(s) in theory and it will answer all questions with a completion. You could ask GPT-2 how will I get into the bank and exit safety and ya it may not get you but it can generate the rest of the story based on a set of conditions (one attempt of breaking in) to see what would happen next. DALL-E turns on all qualia hypermeters, it is breakfest and dinner. ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/T5d9be03603e8ba20-M72755dd71b200e931a10b55f Delivery options: https://agi.topicbox.com/groups/agi/subscription
