OpenAI Goes For-Profit, AOL Gets Bought, AI Marketing (Slop?), and Job Shocks

Author

Bart Smeets
November 07, 2025

Hi all! đź‘‹

A busy week in AI & data news. Check out below for the highlights. The first six we discussed on this weeks pod 👇

Create on-brand marketing content for your business with Pomelli | https://blog.google/technology/google-labs/pomelli/ 

Google Labs and DeepMind launched Pomelli, an AI tool that builds a “Business DNA” from a company’s website to auto-generate on-brand social campaigns. It’s rolling out as a public beta in the U.S., Canada, Australia, and New Zealand—promising speed, but raising questions about originality and disclosure.

OpenAI’s Less-Flashy Rival Might Have a Better Business Model | https://www.wsj.com/tech/ai/anthropic-business-model-ai-9e26b4ef 

WSJ argues Anthropic’s enterprise‑first focus is paying off as corporate customers drive most revenue and adoption. A reported multibillion run rate and strong share in code generation sharpen the contrast with OpenAI’s mass‑market push, setting up a showdown over who monetizes AI best.

Microsoft, OpenAI reach deal removing fundraising constraints for ChatGPT maker | https://www.reuters.com/business/microsoft-openai-reach-new-deal-allow-openai-restructure-2025-10-28/ 

OpenAI is restructuring into a public benefit corporation with Microsoft retaining a ~27% stake, clearing the way for larger capital raises and possible IPO plans. The pact resets rights and revenue‑sharing through 2032, signaling massive data‑center ambitions and tighter alignment amid rising AI costs.

AOL to be sold to Bending Spoons for roughly $1.5B | https://www.axios.com/2025/10/29/aol-bending-spoons-deal 

AOL is changing hands again, with Bending Spoons buying it for about $1.5B as the Italian firm keeps scooping up legacy internet properties. The deal includes ~$2.8B in new debt financing and cites ~30M monthly active users—big numbers for a storied brand.

Accountable Acceleration: Gen AI Fast-Tracks Into the Enterprise | https://knowledge.wharton.upenn.edu/special-report/2025-ai-adoption-report/ 

Wharton’s latest study finds generative AI moving from pilots to measured ROI: most leaders now use it weekly, many daily; 72% track ROI and about three‑quarters report positive returns. The next phase shifts to scaling, talent, governance, and converting adoption into durable performance advantages.

I analyzed 180M jobs to see what jobs AI is actually replacing today | https://bloomberry.com/blog/i-analyzed-180m-jobs-to-see-what-jobs-ai-is-actually-replacing-today/ 

A scrape of 180M job postings shows steep drops in creative execution roles (e.g., computer graphic artists and writers) while machine‑learning engineers surged ~40% YoY. Watchlist: medical scribe postings fell ~20%, hinting AI note‑taking tools are biting into entry‑level healthcare admin work.

Airbnb CEO Brian Chesky warned against replacing entry‑level workers with AI, arguing it risks a future leadership vacuum. He urged companies to keep hiring early‑career talent even if AI can do interns’ work, so future managers can develop skills and judgment.

CoreWeave acquires Marimo to unify the generative-AI developer workflow | https://www.coreweave.com/news/coreweave-acquires-marimo-to-unify-the-generative-ai-developer-workflow 

CoreWeave is acquiring Marimo, creator of the open‑source marimo notebook, to bring an AI‑native notebook experience into its cloud. The company says marimo will stay open‑source while it unifies the developer workflow from experimentation to production for generative‑AI apps.

Fortune highlights Fed Chair Jerome Powell’s view that the AI buildout is generating unusually large economic activity, while raising concerns about market concentration, jobs, and bubble risks. The debate centers on whether AI‑driven capex boom differs from the dot‑com era.

Nvidia is reportedly investing $500M–$1B in Poolside as part of a $2B round valuing the AI code‑generation startup around $12B. The wager underscores Nvidia’s push into developer‑focused models and tools as AI spending shifts from experimentation to software productivity.