Innovation Profs - 9/2/2025

Latest Gen AI News

New Gemini update is “bananas”

Google Gemini got a significant upgrade with a new image editing model powered by DeepMind. The tool, codenamed Nano Banana, preserves consistent likenesses (especially for people and pets) even when altering features like clothing or hairstyles.

You can now blend multiple photos, change backgrounds, and apply styles from one image to another—all while keeping the subject unmistakably recognizable. The model supports multi-turn editing, meaning you can make a series of edits while Gemini continues to preserve the rest of the image with each step. All AI-generated images now include both a visible watermark and an invisible SynthID digital watermark to clearly denote their AI origin. The update is live now in the Gemini app, inviting users to experiment with these enhanced editing capabilities. The new model ranks No. 1 in Image Edit Arena.

Claude comes to Chrome browser

Anthropic introduced a research preview of its browser-based AI assistant Claude for Chrome, available as an extension in Google Chrome. This sidebar agent lets users chat with Claude and delegate tasks, while preserving the context of everything happening in the browser. It’s currently being tested by 1,000 users subscribed to Anthropic’s Max plan (which costs between $100 and $200 per month), with a waitlist open for other interested users. Inside the browser, Claude can perform actions such as managing calendar entries, scheduling meetings, drafting emails, handling expense reports, and navigating websites on behalf of users.

AI apps rankings show shift

a16z’s latest ranking of the top 100 consumer AI apps reveals a stabilizing ecosystem, with only 11 new entrants on the web list and 14 on mobile — both down from previous editions. Google made a strong showing, with four products in the rankings: Gemini (#2 overall), AI Studio (#10), NotebookLM (#13), and Google Labs (#39). Gemini is emerging as ChatGPT’s closest rival, drawing nearly half of ChatGPT’s mobile traffic. Grok 4 surged onto the scene, reaching #4 on web and #23 on mobile. The rise of vibe coding platforms like Lovable (#22) signals a shift toward “personal software,” with stunning early retention and revenue. Meanwhile, AI agent Manus entered at #27.

Sign up for Gen AI Boot Camp

Gen AI Boot Camp: Join us in-person at Drake University or virtually for our next Generative AI Boot Camp. This one-day workshop will get you up to speed and using generative AI tools effectively in your work. Today is the last day to get the early bird rate. Sign up by 5 p.m. today.

When: Friday, Sept. 12, 9 a.m. to 3 p.m.
Where: Drake University or on Zoom
Sign up here

Quick Hits

Tool of the week: Nano Banana

This is turning into a Nano Banana-themed issue of the newsletter. We introduced Google’s new image generation tool above. Now let’s take a look at how you can access Gemini 2.5 Flash Image. You have a couple options. You can go to gemini.google.com and choose “🍌 Create images” under the Tools menu. Alternatively, you can use aistudio.google.com and look for “Try Nano Banana” under the What’s new area.

We tried it out for our image of the week below. But we’ve tested it in a couple other areas as well. Professor Snider asked Nano Banana to complete the five things he teaches in his intro to Photoshop class. And below he used it to restore a photo of his mom (with just the prompt “restore this photo”).

AI-generated image of the week

We tested out Google’s new image generation tool Nano Banana (aka Gemini 2.5 Flash Image) and its ability to alter images. This gave Professor Snider the opportunity to test out some new hair styles. Can you guess which one is the real Professor Snider?

Prompt: Give me a long beard and long hair / Give me a blue mohawk

What we found

OpenAI published its official prompting guide to GPT-5 (linked below), but an X user created a cheat sheet.