📈 How are companies using AI?

➜ AI Helps Mattel Cut Costs: Mattel, the well-known toy company, is using AI to improve its analytics, product development, and integration. Another major goal for AI at Mattel is to help save $200 million over three years. Ambitious plans. 

RetailDive

➜ And Klarna’s Competitor Confirms Cost Cutting with AI Is Real Deal: Zilch’s progress on AI and machine learning would unlock “tens of millions of pounds of cost efficiency in the company just in the next couple of years” CEO of UK’s top fintech said. 

CityAM

➜ Estée Lauder Boosts Campaigns with AI: Estée Lauder has created a generative AI chatbot that quickly comes up with new and tailored marketing campaigns using the company’s data. This kind of AI use is becoming very common for digging into data. AI also allows R&D teams at the beauty company to respond to product and ingredient trends more quickly.

DigitalCommerce360

➜ Bank of London Speeds Up Coding with AI: The Bank of London has introduced a generative AI assistant to make coding faster for companies that are testing embedded banking applications via bank's API suite. AI will soon also help directly chat with the bank's API documentation, making things even smoother.

Finextra

➜ IKEA Doubles Down on AI Literacy: Swedish retailer aims at providing AI training to 3,000 workers and 500 leaders across the organization. Part of the initiative are AI Exploration Days for managers to align AI potential with Ikea business priorities.

RetailDive

➜ Using Fake Data for AI at Big Tech Companies: Major tech companies like Microsoft, Google, and Meta are using synthetic data to train their LLMs. Smart move, as this helps them avoid many legal and ethical issues, and the data created by AI can train new versions of LLMs.

Bloomberg

💡 What are the nerds up to?

➜ Building AI-First Products with LLMs. Mike Knoop, co-founder of Zapier, shared publicly what is his advice to the company's AI engineering teams. Seven pieces of advice, to be more exact. Good read.

Mike Knoop on X

➜ AI Burnout Is Here: CNBC spoke with artificial intelligence engineers at top tech companies. TL;DR: they're not happy. They’re complaining about the burnout, rolling out AI tools at breakneck speed just to appease investors rather than to solve problems for end users.

CNBC

➜ Detecting Bitcoin Dirty Money: Blockchain researchers from Elliptic, IBM Watson, and MIT have used AI to spot money laundering on the Bitcoin blockchain. They claim AI methods help identify illicit wallets and money laundering patterns, which were previously hidden from view.

Finextra

➜ Prompts Behind AI Voice. A month ago I wrote about Hume.ai, the first AI voice capable of reading facial expressions and voice tone for personalized responses. Someone got around guards, and got the original prompt behind Hume. Guidelines are worth checking.

X

➜ LLM Agent Trains Robot Skills. Getting a robot to walk on a yoga ball might not seem like a huge step in AI development, but it is, as it is not possible to accurately simulate the ball’s surface. DrEureka is an LLM agent that writes code to train robot skills in simulation, and writes more code to bridge the difficult simulation-reality gap. 

Eureka

➜ Rapid Prototyping In Hours Instead of Weeks? Shipaifast is a new startup that helps people quickly set up the code for prototyping their ideas and then turn them into AI apps. Founders mix Next.js & Supabase boilerplate with some prebuilt AI apps.

Shipaifast

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.