GenAI Studio: News, Tools, and Teaching & Learning FAQs
These sixty minute, weekly sessions – facilitated by Technologists and Pedagogy Experts from the CTLT – are designed for faculty and staff at UBC who are using, or thinking about using, Generative AI tools as part of their teaching, researching, or daily work. Each week we discuss the news of the week, highlight a specific tool for use within teaching and learning, and then hold a question and answer session for attendees.
They run on Zoom every Wednesday from 1pm – 2pm and you can register for upcoming events on the CTLT Events Website.
News of the Week
Each week we discuss several new items that happened in the Generative AI space over the past 7 days. There’s usually a flood of new AI-adjacent news every week – as this industry is moving so fast – so we highlight news articles which are relevant to the UBC community.
This week, scientists at the University of Graz help bridge the gap between old and new solar data. OpenAI Ceo Sam Altman reveals that polite ChatGPT users are costing the company millions of dollars through the addition of words like “please” and “thank you” in their prompts. Nari Labs introduces a new text to speech dialogue model called Dia-1.6B, able to generate highly realistic dialogue from a transcript. Notable features include the ability to handle multi-modal inputs and maintain consistent voices, even feigning actions like coughing. Stanford researchers present the new tool FramePack, a generative video model that generates video progressively, allowing for performance even on laptop GPUs. Finally, a user on GitHub shares various system prompts from leading AI tools, providing some insight on how AI agents preface your requests to the LLM. By analyzing these system prompts, we can then develop strategies for effective prompting.
Here’s this week’s news:
AI is Helping Scientists Unlock the Secrets of the Sun
Researchers have developed a new AI method to harmonize and enhance decades of solar observations from different instruments. This approach allows data from old and new solar telescopes to be translated into a common scientific language, enabling a more coherent study of the sun’s behaviour over time. By correcting resolution differences and noise, the AI system can improve older images without losing key solar features, enhancing studies of sunspots and magnetic activity. This technique could significantly boost long-term solar research and space weather forecasting. View the entire article here.
Sam Altman On ChatGPT Politeness Costs
Sam Altman, CEO of OpenAI, disclosed that users saying “please” and “thank you” to ChatGPT is costing the company tens of millions of dollars in computing resources. Industry experts note that polite prompts can actually shape more respectful and clear AI responses, enhancing user experience as the chat bot tends to mirror the tone and structure of the given prompts. However, this behaviour raises broader concerns about the environmental costs of frequent and verbose AI interactions. Learn more about the environmental concerns this poses.
Nari Labs Tool Dia-1.6B Demonstrates Realistic Audio Dialogue Generation
Dia-1.6B is a text-to-speech (TTS) model developed by Nari Labs that can generate ultra-realistic dialogue, complete with nonverbal cues such as laughter and coughing. The model allows users to condition speech output on audio samples, enabling precise control over tone and emotion. It also supports dialogue tagging and includes voice cloning capabilities, enhancing realism and speaker consistency. The open-source model is accessible via GitHub and Hugging Face, offering tools and examples for developers to integrate and experiment. See how Dia-1.6B compares versus other models.
Stanford Researchers Release FramePack, Enhancing Video Generation Efficiency
FramePack is a next-frame prediction model developed to make video diffusion more practical and efficient, especially for long video sequences. The model compresses input frame contexts to a fixed length, ensuring consistent computational load regardless of video length. This design enables the use of larger batch sizes and stable processing even on lower-end GPUs, such as laptop hardware. It provides an optimized way to handle video generation tasks with minimal memory and performance bottlenecks. Explore the research behind FramePack.
GitHub User Shares System Prompts for Various AI Agents
Cursor is an AI-enhanced integrated development environment (IDE) assistant designed to streamline software development tasks like coding, debugging, and answering technical questions. Integrated directly into the IDE environment, Cursor uses contextual cues from the user’s activity—such as open files and cursor location—to deliver precise and relevant support. It leverages large language models to act as a real-time programming partner, adapting dynamically to the task at hand. This tool is especially useful for boosting productivity and reducing the time spent on routine development activities. Read the system prompts.
Tool of The Week: UBC GenAI Toolkit (TypeScript)
The UBC GenAI Toolkit is a modular TypeScript library developed to streamline the integration of Generative AI capabilities into web applications. It provides standardized interfaces for common GenAI tasks, abstracting the complexities of underlying implementations. By following the Facade pattern, the toolkit ensures API stability and simplifies the adoption of new technologies, allowing developers to focus on application logic rather than GenAI infrastructure. Access the toolkit.
Questions and Answers
Each studio ends with a question and answer session whereby attendees can ask questions of the pedagogy experts and technologists who facilitate the sessions. We have published a full FAQ section on this site. If you have other questions about GenAI usage, please get in touch.
-
Assessment Design using Generative AI
Generative AI is reshaping assessment design, requiring faculty to adapt assignments to maintain academic integrity. The GENAI Assessment Scale guides AI use in coursework, from study aids to full collaboration, helping educators create assessments that balance AI integration with skill development, fostering critical thinking and fairness in learning.
-
How can I use GenAI in my course?
In education, the integration of GenAI offers a multitude of applications within your courses. Presented is a detailed table categorizing various use cases, outlining the specific roles they play, their pedagogical benefits, and potential risks associated with their implementation. A Complete Breakdown of each use case and the original image can be found here. At […]