Trent Gillespie
December 18, 2025
If you’ve heard me speak, you may know this story. If not, here’s the short version of why I left Amazon.
For nine years, I helped drive Amazon’s global expansion and innovation—from international market entry to Last Mile delivery to Alexa Privacy. I worked on the systems and strategies that let Amazon enter markets, scale fast, and win consistently—with AI embedded deeply into how decisions were made and work got done.
By year eight, a harder question started bothering me:
What happens to the businesses in the communities Amazon enters—the ones that can’t compete with its scale, data, and operational discipline?
Then COVID hit. And it became clear I didn’t want to spend the rest of my career making a massive company bigger for its own sake.
I wanted to take what actually worked—the innovation frameworks, the way Amazon operationalized AI, the systems thinking that made it so effective—and make it accessible to any company willing to do the work, regardless of size.
Not just Fortune 500s.
Not just tech companies.
But companies that know their customers by name, invest in their people, and want to build something that lasts in their communities.
Since leaving Amazon, I’ve worked with leaders across the country on how to operationalize AI—moving past experimentation into real, repeatable productivity. I’ve seen what happens when teams get access to the right tools and the right systems: they don’t just keep up. They pull ahead.
The problem is scale. I’m one person, and the demand is overwhelming.
That’s why I built Stellis AI: to make these frameworks, training methods, and implementation strategies accessible broadly, so more businesses can use AI to grow, not get squeezed. Use it to create jobs, not destroy them.
Now I need your help.
I'm scaling Stellis in 2026 to help more SMBs operationalize AI. But I need to know where you need help most.
Take 30 seconds and vote on this one question:
Your answer directly shapes what we build in 2026. AI agents are already in motion—this input helps us focus on what else will drive the most meaningful impact for companies like yours.
Now, here’s what happened in AI this week—and why it matters less than you think.
Two weeks ago, I shared how Google’s Gemini was beating OpenAI. Last week, OpenAI rushed two updates in response: GPT-5.2 and a major image generation upgrade.
The headlines called it a “huge leap forward.”
The reality is more modest.
GPT-5.2 is focused on science and math, improves spreadsheet and presentation generation, strengthens tool-calling for multi-step workflows, and significantly boosts abstract reasoning. Coding performance also improved.
The image generator is genuinely better—roughly 4× faster, stronger text rendering, and better instruction-following for edits. If you create marketing or visual content, this upgrade matters.
Good updates? Yes.
Revolutionary? No.
When Gemini 3 launched last month, it briefly took the lead, scoring 73 on Artificial Analysis’s Intelligence Index versus GPT-5.1’s 70—the first time Google held the top spot.
GPT-5.2 has now caught up. The models are effectively tied. Google’s Nano Banana image creator is still much better than OpenAI’s latest.
Here’s the point: this race will keep producing improvements—but it won’t determine which businesses win.
While AI labs fight over benchmarks, a more important gap is widening: the gap between companies that know how to use AI effectively and those that don't.
The most effective AI users engage 6× more frequently
They save 40–60 minutes per day
Their productivity gains compound to up to 5× higher than typical users
That advantage doesn't come from choosing GPT-5.2 over Gemini 3.
It comes from three things:
Daily habits beat sporadic experimentation. Master a handful of core use cases—writing, research, analysis, meeting prep—before chasing advanced workflows. The 6× users aren't doing more types of tasks. They're doing the right tasks consistently.
Capability beats subscriptions. AI-capable teams outperform regardless of the model. Teams without training won't extract value from even the best tools. One of our Jumpstart clients delivered a client-ready strategic update this week, just days after launching AI. It took an employee 30 minutes—work that previously took several hours. The client's response: "You seem to know us incredibly well." That's not access to ChatGPT. That's knowing how to extract insight, structure thinking, and deliver client-ready work.
Starting in Q1 compounds all year. Teams that start Q1 with working AI habits will be 5× more productive by December. Teams that wait until Q3 will spend the rest of the year catching up.
The models will keep improving. GPT-6, Gemini 4, Claude 5—they're all coming. But model upgrades won't save businesses that haven't built the foundation.
The technology is ready. The question is whether your organization is.
AI Jumpstart Program – 2 January Spots Remaining
Our 30-day AI Jumpstart takes teams from experimentation to daily, reliable AI use. We align leadership, train teams, and build workflows that actually stick.
For most companies, this is the highest-ROI AI investment they’ll make all year, because it establishes habits and workflows that compound month after month.
If you want your organization AI-capable before Q2, reply now. We’ll assess what you’ve already done, identify the gaps, and help you execute a focused, company-wide rollout, fast.
January Event: How SMBs Win with AI in 2026
I’m hosting a focused January session January 15th, at 10:00 am Pacific, for SMB leaders on what’s actually happening with AI adoption—and what to do about it next.
In this session we’ll break down:
What’s working right now at companies under 500 employees
Where most AI efforts stall or quietly fail
The few moves that matter most in Q1 to build real momentum
This isn’t a product demo or trend recap. It’s a practical, data-backed look at the AI adoption gap—and how the best SMBs are closing it. With open Q&A.
Space is limited. Click here to register.
The reason I left Amazon was to help leaders and companies like yours.
Take 30 seconds on that survey at the top and let me know what type of help would move the needle most for you right now.
If those options don't capture it, hit reply and tell me what does.
I read every response.
About Trent: Trent Gillespie is an AI Keynote Speaker, CEO of Stellis AI, former Amazon leader, and advisor on building AI-Native, AI-Enabled businesses. Book Trent to speak to your group or book a call to discuss using AI within your business.
Tell us what you thought of today's email. |
Did someone forward this newsletter to you? If you're not already signed up, you can subscribe to AI SPRINT™ for free here.