OpenAI’s new 4o Image Generation tool, built into GPT-4o, enables fast, high-quality, photorealistic image creation directly from text prompts, offering better accuracy, native text rendering, and integration with platforms like Adobe and Figma
Generate custom visuals like historical scene recreations, science diagrams, classroom mascots, storybook characters, or visual writing prompts to support lessons and student engagement
Viral uses of 4o image generation include AI-generated yearbook photos, Pixar-style portraits, fake travel selfies, surreal “dreamscape” collages, custom action figure packaging, and mashups like Muppet politicians or medieval fast food ads
Runs on a hybrid ChatGPT 4.0, even in the free model. ChatGPT free will only offer a few 4.0 queries every few hours.
Copilot has Copilot Voice, which recently moved to unlimited use; however, it does not have the option to send a picture or open up the video camera and interact with it.
Copilot will list its sources when it finds any; ChatGPT free will, at some point, start using ChatGPT 3.5, which does not include reference resources.
Copilot has no memory
Google LearnLM
LearnLM is a new family of generative AI models fine-tuned for learning based on Gemini
LearnLM enhances Google products like Search, YouTube, and Gemini, providing deeper understanding and personalized learning experiences
Based on learning science principles, LearnLM supports active learning, manages cognitive load, adapts to learners, stimulates curiosity, and deepens metacognition
Already being used in Google Search, NotebookLM and Google Classroom (Premium)