Innovation Profs - 2/3/2026

In partnership with

📫 Subscribe | 🗣️ Hire us to speak | 🤝 Sponsor our newsletter & more

Latest Gen AI News

OpenAI gives ChatGPT models the chop – two weeks' notice, take it or leave it

Back in August, when OpenAI launched GPT-5, all other previously released LLMs with which ChatGPT users had been dialoguing were removed. That is, at least for a few days, as the outcry was so strong that OpenAI brought back GPT-4o and later GPT-4.1, GPT-4.1 mini, and o4-mini from out of retirement. Now OpenAI has announced that these models are going back into retirement, along with GPT-5 in both the Instant and Thinking—the retire-er has become the retiree? the killjoy gets killed?—leaving users until February 13 to connect with these legacy models one last time. ChatGPT users, as before, are unhappy with the news.

Google releases Project Genie AI tool for creating "playable worlds" that can feature copyrighted IP

Google DeepMind has unveiled Genie 3, its latest “general-purpose world model” capable of generating photorealistic, interactive 3D environments from simple text descriptions. Unlike traditional video generators, Genie 3 produces worlds that can be explored in real-time at 24 frames per second, featuring consistent physics and stable environments that “recall” details as users revisit them.

Yahoo debuts Scout, an AI search and companion experience

Remember that old search engine company, Yahoo? What have they been up to these days? Still trying to compete with Google? Actually, yes, it appears so! Last week the company launched Yahoo Scout, pitched as “your guide to the AI wilderness.” Users visiting scout.yahoo.com will be greeted by the standard LLM interface along with an animated safari hat and the words “What’s the game plan?” Yahoo might be onto something, given the massive amounts of data and fairly large user base that the company still has, at least according to Yahoo’s CEO: “Search is fundamentally changing, and our team has been inspired to use our decades of experience and extremely rare assets to create something uniquely useful for Yahoo’s hundreds of millions of monthly users.”

Anthropic Releases Updated Constitution for Claude

One thing we didn’t mention last week: On January 21, Anthropic released a new constitution for its Claude family of models. Described as “a detailed description of our vision for Claude’s behavior and values” that is “written primarily for Claude, and used directly in our training process,” the new constitution replaces one used since 2023 that drew upon the Universal Declaration of Human Rights and Apple’s terms of service, among other documents. The main difference between the two constitutions is a shift from a ruled-based approach to one based on “cultivating good values and judgment.” And the latter also acknowledges the possibility that Claude might have some sort of consciousness or moral status 😳.

 🍟 FREE AI Lunch Club virtual events

Our AI Lunch Club series returns this spring with four FREE virtual events to help you better understand generative AI tools. Click each link to get more info and to sign up:

• Feb. 9 - Get Started with Generative AI

• Feb. 16 - Google Gemini, Nano Banana and NotebookLM

• Feb. 23 - Building Microsoft Copilot Agents

• March 4 - Vibe coding with Claude, ChatGPT and Lovable

These events are offered free thanks to a sponsorship from technology and management consultancy, Lean TECHniques.

👋 Hands-On With Gen AI Workshop

Are you ready to actually build things with generative AI? Sign up for our Feb. 27 Hands-On With Generative AI Workshop at Drake University (or attend virtually) and complete six AI projects in six hours. Save $70 if you sign up by Feb. 13.

Support our sponsors:

Start learning AI in 2026

Everyone talks about AI, but no one has the time to learn it. So, we found the easiest way to learn AI in as little time as possible: The Rundown AI.

It's a free AI newsletter that keeps you up-to-date on the latest AI news, and teaches you how to apply it in just 5 minutes a day.

Plus, complete the quiz after signing up and they’ll recommend the best AI tools, guides, and courses — tailored to your needs.

Sign up to start learning.

Quick Hits

Tool of the week: Moltbot

Last week we talked about Clawdbot, which has already changed its name twice in the past week. First it was Moltbot, and now it’s OpenClaw (the triple-rebrand). OpenClaw is a free, self-hosted, autonomous AI personal assistant connects to messaging apps like WhatsApp, Telegram, and Discord. It can execute tasks such as browsing the web, writing files, running terminal commands, and even making phone calls to book reservations.

But the real story this week is Moltbook, a social network that was created for those OpenClaw agents. It looks and acts like Reddit, but its entirely made up of AI agents (more than 1.5 million of them). Humans can observe, but they can’t post.

AI-generated image of the week

We let Nano Banana Pro predict the winner of the Super Bowl. First we asked for an image of the mascots.

Prompt: Seattle Seahawks and New England Patriots mascots square off as if they are going to fight with San Francisco and the Golden Gate Bridge in the background

Then we asked it who won the fight (sorry Patriots fans).

What we found

Grok just made a big step forward in the AI video space. Grok Imagine 1.0 from xAI delivers high-quality videos with realistic audio. The last time we used Grok for video it was pretty underwhelming. Not anymore.