Welcome to Jumble, your go-to source for AI news updates. This week, Ford admits its AI quality push backfired and rehired hundreds of veteran engineers, while South Korea bets a record $576 billion on the chips behind every AI. Let’s dive in ⬇️

In today’s newsletter:
👴 Ford rehires the engineers AI couldn't replace
📟 South Korea stakes a record fortune on chips
🧠 OpenAI teases its next flagship model
🛰️ Rocket Lab makes a play to run all of space
🕵️ Weekly Challenge: Put your AI through quality control

🔩 Ford Rehired the Engineers Its AI Couldn't Replace

Ford trusted AI and automated systems to catch its quality problems for years, and it flopped. After topping the latest JD Power quality survey, the company admitted the real fix was human.

Are companies finally realizing that AI still needs a human in the loop?

Login or Subscribe to participate

👷 Bringing Back the Vets

Ford rehired 350 veteran engineers, many of them former staff, to train younger workers and reprogram the AI tools that kept missing the mark.

🤔 The AI Was Only as Good as Its Training

A Ford engineering lead said they "mistakenly thought that by just introducing artificial intelligence" they would get a quality product. The veterans now catch failures before a part reaches the line.

💰 The Human Touch Is Paying Off

Ford expects the move to drive roughly $1 billion in lower costs this year. It isn't ditching AI, just pairing it with people who know where the bodies are buried.

🌏 South Korea Bet $576 Billion to Own the AI Chip Race

President Lee Jae Myung unveiled more than $576 billion in chip and AI investment, anchored by Samsung and SK Hynix. He called semiconductors, physical AI, and data centers the country's "triple axis."

🏭 Two Brand-New Fabs Each

Samsung and SK Hynix will spend 800 trillion won with suppliers on two new chip plants apiece in the southwest. The aim is overwhelming capacity before demand outruns it.

📉 A Huge Bet With Real Risk

Both chipmakers' shares slipped as some analysts warned of a possible supply glut. But their grip on the memory inside modern AI makes them hard to bet against.

Weekly Scoop 🍦

🔎 Weekly Challenge: Put Your AI Through Quality Control

Challenge: Ford learned AI needs a human checking its work. This week, be the inspector and see whether ChatGPT, Gemini, and Claude get the facts right on a topic you know cold.

Here's what to do:

📝 Step 1: Pick your specialty. Choose something you know well, like your job, hometown, or a hobby. You're grading the answers.

🔁 Step 2: Ask all three. Give the same detailed question to ChatGPT, Gemini, and Claude and compare the answers side by side.

🚩 Step 3: Catch the bluffs. Flag confident claims that are flat wrong, then ask each one for a real, checkable source.

🏆 Step 4: Crown a winner. Reply with which AI was most accurate, and the most ridiculous thing one made up.

Would you rather an AI or a 30-year veteran sign off on your next car? And is South Korea's $576 billion bet visionary, or about to flood the world with chips? See you next time! 🚀

Stay informed, stay curious, and stay ahead with Jumble!

Zoe from Jumble

Keep Reading