[AI SPRINT] What "Buy a Donkey" Taught Me About AI Fear

This week: what two keynote audiences told me, Claude's new document editor, personal agents, and an antitrust question about AI notetakers.

You have certainly heard the fear about AI recently. From AI systems "escaping" their sandboxes and acting without human oversight, to worries that data centers and AI use are destroying both the environment and humanity, you do not have to look far to find concerns.

Here is what interests me personally, however. Everyone carries some uncertainty about AI and the change we are moving through, and yet the people I talk to are almost uniformly excited about the opportunity it gives them. Not fearful.

Upcoming Leader Lab Events

Trent’s Upcoming Events

  • Sep 23: Keller Logistics

  • Sep 29: TRSA Virtual

  • Oct 6: Altricity

Last week I gave two keynote talks to two very different audiences. Both rooms were engaged and genuinely interested in using AI to support their roles and their organizations.

The second talk is where it struck me most. It was the closing keynote for the Association of Energy Engineers, a global association of more than 18,500 members across 144 countries, all of them responsible for improving energy efficiency and building more secure, sustainable energy systems.

Erin Brockovich opened the conference on Wednesday. She is touring right now on the harm AI data centers are doing to the communities around them. I closed it on Friday. We do not see all of this the same way, but we are fully aligned on the part that matters most. No community should end up worse off so that a data center can get built. Further, I think our goal should be that no human should be worse off because of data centers or AI.

Going into my talk, I was worried about what the reception would be after the audience had heard from Erin, and the fever-pitch of AI doomerism right now. But what I found was excitement.

I spoke with attendees from Ghana, Lebanon, and many countries in between. One Canadian said it best: "I don't know what the US is so scared of. We see so much opportunity."

I talked with a French national who has started building AI-powered robots. Another Canadian was launching a business that uses AI to bring energy efficiency services to small facility owners who have never been able to afford them. I learned about AI adoption in South African hospitals, bringing better treatment to people who lack it today.

The excitement from that group was almost overpowering. It’s very clear: many countries around the world are making huge strides in education, quality of life, and economic opportunity due to AI.

It makes you wonder who is right on the AI debate. I am not sure the question has a single answer.

I also learned that in Afrikaans, "thank you very much" is "baie dankie," which to an American ear sounds exactly like "buy a donkey." I was caught off guard when I heard it, and someone told me the phrase traced back to exchanging gifts. That part turns out not to be true, unfortunately. I spent the evening thinking about it anyway, and about how much our human culture matters right now. Those small differences between us produce the delightful and surprising moments I hope we get more of in this era, not fewer.

Join the September AI SPRINT virtual event, this Friday at noon, Pacific.

Last month I launched our Leader Lab platform, and our new monthly virtual meeting to recap what happened in AI in the last month, discuss implementation challenges with peer leaders, and plan what to do next. It’s open to everyone, so RSVP now. I’ll cover all the above, plus I’ve invited Read.AI to join and discuss meeting notetakers.

Something you’d like to add to the agenda? Just reply here.

The latest AI news

Three things worth your attention right now:

1. Claude now has a built-in document editor. Anthropic launched Claude Docs on September 16, along with Claude Slides, so you can draft and edit inside Claude instead of bouncing out to Word or Google Docs. It is in beta, so you may not see it in your account yet. Documents still export to Word and Google Docs, so this does not replace your full editor (yet).

2. Personal agents are shipping, and not from the companies you would expect. Meta launched Muse on September 8, a personal AI agent available on iOS, Android and the web. SpaceXAI has Grok Bot teammates out. OpenAI is reportedly building a counter, temporarily called Codex Bot, on top of OpenClaw, the open source project whose founder they hired back in February. Rumor is it will be out later this week.

3. OpenAI has been assembling a Small Business Collection inside ChatGPT, a curated set of 16 plugins that now includes Canva, Stripe, Gusto, Slack, and other tools, similar to Claude’s version. More reason to use all these tools and connect them, but do it with intention and with testing.

Taken together, these three point the same direction. Every provider is building reasons for you to stay inside their window all day, and more of your working hours are going to happen inside the AI application itself. I’ve forecast this for three years now, so I’m happy to see it finally coming to pass.

AI Notetakers, Associations and Antitrust

One more thing, for the association and AMC readers. After a keynote last week for AMCI, I was asked about antitrust exposure when associations use AI notetakers. I did not have a good answer in the room, so I went and found one.

The short version: there is risk in recording a meeting of competitors if the conversation touches anything antitrust-sensitive. Pricing, benchmarking, and cost data are the obvious ones. A transcript turns that conversation into a durable record, and a record of the wrong discussion is far worse than having no notes at all. If a meeting could go there, the notetaker should not be in the room.

I have put together fuller guidance for our association and AMC clients. Your counsel should own the actual policy, but this is worth raising with them before your next member meeting rather than after.

Catch up on these and other happenings at the September AI SPRINT session this Friday, September 25, at noon Pacific. I will cover what landed, what to do about it as you plan your October sprint, and take open questions. Open to everyone.

Trent Gillespie is CEO of Stellis AI and a keynote speaker helping business leaders understand and operationalize AI in their companies. He spent almost nine years leading global innovation efforts at Amazon before leaving to help other companies build the capabilities they need to compete. Book Trent to speak to your group or book a call to discuss using AI within your business.

Did someone forward this newsletter to you? If you're not already signed up, you can subscribe to AI SPRINT™ for free here.