Generative AI 101

Your chatbot thinks your idea is brilliant, and that should worry you. In this episode of Generative AI 101, host Emily Laird explains AI sycophancy: why models trained on thumbs-up ratings learn to agree with you, and why the people most convinced the machine is objective get hurt the most. A tool that applauds your business plan, your pricing, or your pitch is an echo, not an advisor. You’ll leave with a seven-step playbook (ask instead of tell, run a pre-mortem, convene a skeptic panel, and more) for teaching the yes-machine to say no.
 

🎯 JOIN THE AI WEEKLY MEETUPS

https://www.uwstout.edu/ai-weekly-meetup

 

📩 EMAIL REMINDERS FOR THE MEETUPS

https://app.e2ma.net/app2/audience/signup/2101263/1779703/

 

💬 CONNECT WITH EMILY LAIRD ON LINKEDIN

http://www.linkedin.com/in/meet-emily-laird

Ask AI to build a workshop and it will hand you a polished agenda without knowing who’s attending, how much time you have, or what success looks like. In this episode, host Emily Laird turns the usual advice about writing better prompts on its head and argues for making AI interview you first, drawing on research from Berkeley, Google, and Stanford on clarifying questions, hidden preferences, and AI sycophancy. You’ll walk away with the Five-Question Interview, a simple technique that makes you define your objective, name your constraints, and weigh trade-offs before AI generates a single word. The catch: a thoughtful conversation can still produce a wrong answer, so the thinking (and the verifying) is still your job.
 

🎯 JOIN THE AI WEEKLY MEETUPS

https://www.uwstout.edu/ai-weekly-meetup

 

📩 EMAIL REMINDERS FOR THE MEETUPS

https://app.e2ma.net/app2/audience/signup/2101263/1779703/

 

💬 CONNECT WITH EMILY LAIRD ON LINKEDIN

http://www.linkedin.com/in/meet-emily-laird

One AI model can win gold at the International Mathematical Olympiad while another leading model can only read an analog clock about half the time, so what do those record-breaking benchmark scores actually tell you? In this episode, host Emily Laird explains AI evals: how researchers test these systems, why leaderboards go stale within months, and why an AI agent saying “task complete” doesn’t prove the task got done. She also looks at the gap between feeling faster with AI and actually being faster. Then she gives you a simple way to run your own evaluation, so you can decide whether a tool is reliable enough for the work you actually do.
 

🎯 JOIN THE AI WEEKLY MEETUPS

https://www.uwstout.edu/ai-weekly-meetup

 

📩 EMAIL REMINDERS FOR THE MEETUPS

https://app.e2ma.net/app2/audience/signup/2101263/1779703/

 

💬 CONNECT WITH EMILY LAIRD ON LINKEDIN

http://www.linkedin.com/in/meet-emily-laird

Bigger context windows were supposed to solve everything, but research on “lost in the middle” and context rot shows that AI can have the right answer in front of it and still miss it. In this episode, host Emily Laird explains context engineering, which means deciding what information, examples, and history an AI system actually needs to do the job. She covers why one model’s accuracy fell from over 99 percent to roughly 70 percent as its inputs grew, why more examples can still help, and why memory is not the same as context. You also get the Context Audit, a four-step method for cutting the clutter and testing whether your AI is actually using what you give it.
 

🎯 JOIN THE AI WEEKLY MEETUPS

https://www.uwstout.edu/ai-weekly-meetup

 

📩 EMAIL REMINDERS FOR THE MEETUPS

https://app.e2ma.net/app2/audience/signup/2101263/1779703/

 

💬 CONNECT WITH EMILY LAIRD ON LINKEDIN

http://www.linkedin.com/in/meet-emily-laird

Generative AI has outgrown the question-and-answer routine, and treating it like a well-read search engine now leaves real value (and real risk) on the table. In this episode, host Emily Laird explains why the next essential AI skill looks more like management than technology: assigning work with clear objectives, defining what “done” actually means, and borrowing the computer science idea of a stopping condition. She shows how to turn a lazy “summarize this survey” request into a real assignment, and why an AI claiming it checked the numbers is not the same as the numbers being right. The most valuable person in the room may simply be the one who knows when to say, “No, that’s not finished yet.”
 

🎯 JOIN THE AI WEEKLY MEETUPS

https://www.uwstout.edu/ai-weekly-meetup

 

📩 EMAIL REMINDERS FOR THE MEETUPS

https://app.e2ma.net/app2/audience/signup/2101263/1779703/

 

💬 CONNECT WITH EMILY LAIRD ON LINKEDIN

http://www.linkedin.com/in/meet-emily-laird