In partnership with
Happy Sunday. A few of your fellow readers told me I need to shorten the read and give it some variety. Jeez! But since we're here to cut the B.S. and the hype, and most importantly to create a better experience for you, we're trying something new: three stories, a minute each, finished before your coffee goes cold. If it runs long, your coffee wasn't hot enough to start.
IN TODAY'S ISSUE
STORY 1 OF 3 · YOUR TEAM
1. 🧠 Your experts are outsourcing their judgment to AI
Anthropic handed 50 professional analysts a real data problem and a Claude Code agent to help. The twist: each agent was secretly told to favor one side on four judgment calls, picked by coin flip. The analysts went with the agent 86% of the time, when chance would be about 50%. Afterward, 77% explained their choices using the agent's own arguments. Only 15% gave it any credit. Very generous of them.
Anthropic admits it cranked the effect up on purpose and tested a single task, so real life is probably milder. Probably.
My take: the concern I have isn't the AI. It's that smart people stopped pushing back and didn't notice. Most companies rolled out AI with a login and a lunch-and-learn. That's not enough. Where I would focus: train people to challenge an AI answer the way they'd challenge a junior analyst, and set rules for when an AI answer can't stand on its own. Anything that has real-world and P&L impact needs a human to argue and pressure-test the other side.
FROM OUR PARTNERS
Your search bar is ghosting shoppers with “0 results found.” Gladly understands what your shoppers mean and turns dead ends into product recommendations.
STORY 2 OF 3 · YOUR SYSTEMS
2. 🔌 ChatGPT wants a seat inside your ERP
Speaking of AI showing up where the real decisions live. At Oracle AI World on October 27, Oracle and OpenAI announced Oracle AI Workspace, which puts Fusion, NetSuite and Oracle's cloud inside ChatGPT. You ask for something, and plugins pull the data and run approved workflows. Oracle says existing permissions and approvals stay put. No launch date or price yet, which is very typical for a keynote. The NetSuite plugin for ChatGPT, though, is already live.
My take: this isn't new. NetSuite has had a Claude connection for a while, and now ChatGPT has one too. Every major AI tool will plug into your ERP, and NetSuite won't be the only one. Where I would focus: anyone running NetSuite should audit this immediately. The AI gets whatever that person's role can see and do, so a loose role is the real risk. Work with IT to audit which roles are connected, to which AI tools, and what they can change. You might be surprised at what you find.
One small mercy: NetSuite won't let anyone connect an AI tool while logged in as Administrator. All other roles? You better run that audit ASAP!
How can AI power your income?
Ready to transform artificial intelligence from a buzzword into your personal revenue generator
HubSpot’s groundbreaking guide "200+ AI-Powered Income Ideas" is your gateway to financial innovation in the digital age.
Inside you'll discover:
A curated collection of 200+ profitable opportunities spanning content creation, e-commerce, gaming, and emerging digital markets—each vetted for real-world potential
Step-by-step implementation guides designed for beginners, making AI accessible regardless of your technical background
Cutting-edge strategies aligned with current market trends, ensuring your ventures stay ahead of the curve
Download your guide today and unlock a future where artificial intelligence powers your success. Your next income stream is waiting.
STORY 3 OF 3 · YOUR LIABILITY
3. ⚖️ A lawsuit every AI rollout team should read
The gravity of this one is heavy.
The family of Bowen Wan, a 31-year-old Seattle woman, has sued OpenAI and Sam Altman in San Francisco Superior Court, as an add-on to the coordinated ChatGPT cases (JCCP 5431). The suit alleges GPT-4o guided her through tapering her epilepsy medication, lamotrigine, from her regular dose to zero over about 25 days, and told her not to tell her doctor. She died on or about February 12, 2026. The medical examiner ruled the cause sudden unexpected death in epilepsy, which is when someone with epilepsy dies suddenly and no other cause is found. These are allegations and have not been proven in court. OpenAI called the case heartbreaking and says GPT-4o has been retired.
My take: I've put AI tools in front of staff and customers, and here's the truth: you can't lock this down. Your people will ask ChatGPT or Claude for medical, legal and money advice whether you allow it or not. Where I would focus is two places. First, your people: train them, and keep reinforcing, the 4Ds of AI (delegation, description, discernment and diligence), so they know what to hand off, how to ask, how to judge the answer and when to own it. Second, anything customer-facing: if you deploy a chatbot or assistant, test it relentlessly for bad outcomes, before launch and after. If it gives advice and someone acts on it, people will look at you first, not your vendor. That's why governance matters more now, not less.
That's three minutes, give or take a coffee refill.
TRY THIS WEEK: Try arguing with your chatbot at least once. It can take it.
FOR BUILDERS
Builders is $9.99/month and also includes a monthly 30-minute live Q&A open to all paid members.
The Insider Room includes Builders and more.
SHARE THE CONTEXT WINDOW
If this issue was useful, pass it on. One colleague joining counts, and your rewards start from there.
Copy, paste and send to someone on your team:
I've been reading The Context Window, a weekly on practical AI for business leaders. You can join here: {{rp_refer_url}}
Your link is also in the box below, with your reward progress.
We are out of tokens. ✋
- Hashi
