Turns out, even Spotify isn’t immune to AI spam.

Last month, the company wiped 75 million AI-generated tracks from its platform—roughly the size of its entire real music catalog.

Each one was gaming the system: looping 31-second songs to trigger royalties and siphon money from actual artists.

It’s a reminder that when AI meets incentives, the results can get messy fast—and not always in tune.

So please, implement it properly…

With that said, here’s what’s new in the AI world this week:

📈 How are companies using AI?​

➜ Playtime with AI. Mattel is partnering with OpenAI to experiment with Sora-powered video tools and AI-driven play experiences that blend physical toys with generative content. The tie-up could turn storytime into an interactive, personalized media playground.
Reuters

➜ Holiday Ads but Auto-Generated. Toys “R” Us is leaning into AI for marketing and brand films, including early experiments with OpenAI’s text-to-video Sora to produce seasonal creative at speed. Expect more hyper-targeted, snackable holiday content built faster than traditional ad shoots.
Retail Dive

➜ Local Discovery, Amped. Yelp’s fall product release rolled out upgrades aimed at helping users find the right businesses faster—features that lean on automation and smarter recommendations. It’s a push to make discovery more contextual and useful at the moment you need it.
Yelp Blog

➜ Delivery Gets a Heads-Up. Amazon unveiled smart glasses for delivery drivers that surface navigation, package info, and hands-free workflows—designed to speed deliveries while keeping drivers’ eyes where they need to be. It’s an on-the-job AR assist that folds AI into every stop.
TechCrunch

➜ Cars That Converse. GM is building a unified software platform that layers conversational AI into the driving experience and across operations—aiming for smarter in-vehicle assistants and more responsive manufacturing systems. Think voice that actually understands context, not just commands.
GM Newsroom

💡 What are the nerds up to?

➜ Wall Street Meets LLM. OpenAI hired former Wall Street bankers to help train models on financial data and risk workflows, signaling a deeper push into finance-grade AI capabilities. The move hints at models tailored for market nuance and regulatory contexts.
Finextra

➜ Likeness Police. YouTube rolled out or expanded AI tools to detect deepfakes and protect creators’ likenesses—part of a broader effort to spot manipulated media before it spreads. It’s content policing aimed at preserving trust on high-visibility platforms.
The Verge

➜ Celebs, Labs, and Superintelligence. A round of op-eds and interviews this week revisited how early AI pioneers and public figures talk about superintelligence—mixing cautionary takes with calls for clearer governance. The chatter keeps pressure on policy and funding debates.
Morning Brew

➜ Weather Forecast by Model. Researchers are applying machine learning to improve short- and medium-range weather forecasts—cutting model noise and giving cities better lead time for extreme events. Smarter forecasts mean smarter logistics, emergency response, and crop planning.
Spectrum IEEE

➜ B2B Payments Are Getting Intelligent. Intelligent agents are starting to automate B2B payment flows—validating invoices, routing approvals, and reducing friction in cross-company reconciliation. It’s the quiet plumbing that could speed cash cycles and cut costs across industries.
PYMNTS

Thanks for reading!

I share these stories because I believe it's important for all of us to keep up with AI. To support my mission, share this newsletter on LinkedIn.