Innovation Profs - 3/26/2024

Your weekly guide to generative AI tools and news

Sign up for AI Lunch Club

Want to go deeper into the subjects we talk about in our workshops? We’re doing just that in our new AI Lunch Club virtual events. These will take place over Zoom from 12:30-1:30 p.m. Central:

  • April 1: GPTs and Me: Learn to use and make custom GPTs.

  • April 3: Go AI Yourself: Create an avatar and make it talk.

  • April 8: Advanced Image Prompting: Take a deep dive into image creation.

  • April 10: Next Steps in Prompting: Advanced prompting techniques for LLMs.

Sign up for events here. Use the code SAVE10 to save $10 off the price for each event. Or bring 3+ people and save more.

Generative AI News

OpenAI is expected to release a 'materially better' GPT-5 for its chatbot mid-year, sources say

According to several anonymous sources, OpenAI is planning to release the model GPT-5 for ChatGPT this coming summer. Reportedly, some ChatGPT Enterprise users have been given demos of GPT-5. As described by one CEO who was given access to the model, "It's really good, like materially better." One source claimed that the model is still being trained, after which it will be safety tested prior to deployment.

OpenAI demonstrates video-generating system to Hollywood directors

On the video generation front, OpenAI has been demoing it text-to-video tool Sora to unnamed Hollywood film studios and directors. According to an OpenAI spokesperson, "OpenAI has a deliberate strategy of working in collaboration with industry through a process of iterative deployment — rolling out AI advances in phases — in order to ensure safe implementation and to give people an idea of what’s on the horizon." Whether Hollywood embraces this technology remains to be seen, particularly in light of unresolved copyright issues and the potential for significant displacement of film industry workers.

Stability AI CEO resigns because you’re ‘not going to beat centralized AI with more centralized AI’

Stability AI, the company behind the AI image generation tool Stable Diffusion, has had a rough past few weeks. After the departure of three of the original five researchers who developed the platform earlier this month, Emad Mostaque, the founder and CEO of Stability AI, just announced that he was stepping down from his position to pursue work on “decentralized AI.” This comes after reports that the company was burning through cash reserves and struggling to raise investor funding.

Microsoft hires DeepMind co-founder Mustafa Suleyman as CEO of new AI unit

Last week Microsoft announced that they had hired Mustafa Suleyman, one of the co-founders of Google DeepMind and co-founder and current CEO of Inflection AI, to oversee Copilot AI initiatives as CEO of Microsoft AI. The move comes with further moves impacting Inflection AI; see below.

Microsoft pays Inflection $650 mln in licensing deal while poaching top talents, source says

Along with the hire of Inflection AI’s Suleyman, Microsoft will pay Inflection AI $650 million to use Inflection AI’s language models, which will be hosted on Microsoft’s Azure cloud service, and hire most of the 70-person team at Inflection AI, including Karén Simonyan, another Inflection AI co-founder, as Chief Scientist of Microsoft AI. With this move, Microsoft continues to invest heavily in generative AI (alongside its partnership with OpenAI), which will likely draw further attention from US antitrust regulators.

Quick Hits

Tool of the week: Open AI Cookbook

Ready to take your prompting skills to the next level? First, sign up for Professor Porter’s Next Steps in Prompting lunch and learn. Second, check out Open AI’s Cookbook resource on Github.

Click on Articles and then click on related_resources. You’ll find articles, guides and videos to improve your prompting.

AI-generated image of the week

This week has been a tough one for Drake basketball fans. Here’s an image that captures how we feel.

Prompt: a row of sad bulldogs watch a March Madness basketball game, with tears in their eyes, photorealistic.

Generative AI tip of the week: Image prompts

When writing prompts for generative AI images, make sure to avoid ambiguity. Your prompts should be clear and direct to get the results you want. A prompt that just says “cat” leaves a lot for the tool to interpret. “Photo of a grey cat sitting on a picket fence,” is much more clear.

Get starting with Generative AI

New to generative AI? Here are some places to start…

What we found

shy kids, a film-making collective from Toronto, was given access to OpenAI’s text-to-video tool Sora to make short film “air head.” Watch it below. For more short pieces created by visual artists using Sora, check out this recent OpenAI blog post.