Multimodal Generative AI with Leon Furze | teacher resource
Consultant, author and PhD candidate Leon Furze returns for another engaging series for educators, further exploring Generative Artificial Intelligence and its multimodal capabilities.
Following on from Demystifying AI, this series applies ideas relating to:
- exploring new Generative AI models that move beyond text into images, audio, video, and even playable games.
- how Generative AI can enhance film production and its relationship to film, media studies, English and visual arts.
- creating videogames using powerful language models
1: Introduction to Multimodal Generative AI
Since ChatGPT was released in November 2022, generative Artificial Intelligence has come a long way. While we've been trying to keep up with the changes to education brought in by language models, developers have been rolling out technologies that can generate not just text but images, audio, video, and even playable games.
New generative AI models also recognise speech and images, edging them closer and closer to being fully multimodal. This session looks at multimodal generative AI in 2024, reintroducing applications like ChatGPT and Microsoft Copilot, and demonstrating some of the more powerful technologies released in the last six months.
2: Creating Videos with Generative AI
Video generation is still in its early days, even with powerful tools. It's possible to create impressive scenes based on simple text or image prompts. This session looks at how generative AI can be used right now to create videos and explores emerging technologies like OpenAI's powerful Sora model and Google's tools for interpreting video.
This session is perfect for anyone interested in film and media studies, English, visual arts, or anyone interested in creating with generative AI.
3: Generative AI and Video Games
We're fast approaching a world of video games on demand. Generative AI can be used to create characters, levels, and even playable scenes. It can help with game design, asset creation, and code. Alongside advances in virtual reality, augmented reality, and powerful language models like GPT-4, the implications for generative AI and video game design are huge.
This session will run through what's possible right now and what's on the near horizon, including models like Google's Genie (which can generate playable levels from simple text prompts), the incorporation of generative AI tools into existing game design platforms, and technologies from companies like Unreal Engine and Nvidia.