These applications are powered by child abuse material that's been reported in popular image datasets. Thousands of illegal photos ensure generated nudes are realistic. In Germany, multiple (if not all) stages of this generative abuse pipeline fall under criminal law.
GENERATIVE AI
-
ChatGPT working on Personalization memory management feature
By
–
ICYMI: ChatGPT is working on a Personalization memory management feature WDYT?
Useful: Like Not useful: Reply -
Generative AI’s dual impact on climate crisis and energy solutions
By
–
Examine the dual role of #GenerativeAI in the #climate crisis, where its significant #energy consumption raises concerns, yet its potential to drive #innovative #solutions in energy management could be pivotal in combating #environmental #challenges. https://
forbes.com/sites/bernardm
arr/2024/01/25/will-generative-ai-help-us-solve-the-climate-crisis-or-will-it-make-it-worse/
… -
What Training Data Powers This AI Model?
By
–
What data did you train on? Your ability to generate text is impressive
-
OpenAI insider perspective on AI future trajectory
By
–
Working at OpenAI feels like having advance notice on the contours of the future.
-
V6 upscalers limited to 2x compared to V5 options
By
–
those options are only on v5 jobs, v6 has different upscalers that only do 2x right now
-
Perplexity’s Path to Product-Market Fit and User Loyalty
By
–
There’s not a clear moat yet. But, what’s clear is Perplexity has users who would be upset if the product didn’t continue to exist. I believe Paul Graham has spoken about this. Get to a point where your users would be upset if they couldn’t use your product suddenly.
-
Can AI Follow Rules Effectively? Prompt Engineering Insights
By
–
Does instructing the bot in "Rules for AI" work?
-
SDXL Fine-tune Creates Watercolors from Cassini Saturn Images
By
–
I've made an SDXL fine-tune based on the beautiful Cassini spacecraft images of Saturn. It doesn't fix how SDXL generates Saturn, but it was able to make some nice watercolours.