CEO Memo: You Will Use AI
“Reflexive AI usage is now a baseline expectation” – Shopify CEO
The AI Jobs Question
I started writing “AI Changes Everything” to both evangelize the message of the importance of AI and to help people navigate through the AI revolution informed. One of the biggest concerns people have about AI relates to their job, and I have been often asked question around AI displacing human work, including “Will AI take my job?”
When people have asked me about this, I have repeated a line that has become a cliché:
AI won’t take your job; the people who use AI will take your job.
Two years after the ChatGPT moment, AI use has now permeated our lives at work and at home. We use AI for advice, brainstorming, writing, medical questions, brand development, research, report generation, and generating all manner of creative artifacts, from images, videos, and songs. Some are leading adopters, while others trail, but many will be using or interacting with AI even without their knowledge.
In the workplace, AI has gone from toy to tool and from “nice-to-have” to “must-have.” That ‘customer support agent’ online or on the phone might be an AI. All this AI automation leaves many questioning the fate of human jobs. However, the latest memo from Shopify CEO Tobi Lutke makes clear: AI is in the workplace to stay; every employee at Shopify is mandated to use AI in their work.
Shopify’s CEO Writes a Memo
The Shopify founder and CEO Tobi Lutke has been leading Shopify to be a leading e-commerce platform. He recently wrote an internal memo to Shopify employees which got leaked and caused some controversy.
So Lutke just went ahead and published his CEO memo on X. It’s rare to get a glimpse of blunt honesty and corporate strategy from a CEO like this out in public. It reminds me of Bill Gates in 1995 and his “Internet Tidal Wave” memo that signaled Microsoft’s recognition of the importance of the internet.
CEO Lutke knows the importance of AI and has been leaning in on using AI personally and as part of the Shopify set of features for their e-commerce customers. His memo goes further, putting its main message front and center:
Our task here at Shopify is to make our software unquestionably the best canvas on which to develop the best businesses of the future. We do this by keeping everyone cutting edge and bringing all the best tools to bear so our merchants can be more successful than they themselves used to imagine. For that we need to be absolutely ahead.
Reflexive AI usage is now a baseline expectation at Shopify.
He mentioned his own use of AI and the use and tinkering by others at Shopify to accelerate their progress, learning that “AI acts as a multiplier.” His goal in the memo is to harness that power of AI to advance their business.
While the memo couches this opportunity to leverage AI in terms of corporate optimism and credos like “core values of Be a Constant Learner and Thrive on Change,” the message is blunt and clear: Shopify needs their team to use AI to get more productive and grow. Using AI is not just an opportunity, it’s a necessity.
The AI Use Expectations
The memo centers around six expectations CEO Lutke set for Shopify employees around AI use:
Using AI effectively is now a fundamental expectation of everyone at Shopify. It's a tool of all trades today and will only grow in importance …
AI must be part of your GSD Prototype phase … Prototypes are meant for learning and creating information. AI dramatically accelerates this process. …
We will add AI usage questions to our performance and peer review questionnaire. Learning to use AI well is an unobvious skill. …
Learning is self-directed but share what you learned. …We’ll learn and adapt together as a team. …
Before asking for more Headcount and resources, teams must demonstrate why they cannot get what they want done using AI. What would this area look like if autonomous AI agents were already part of the team? …
Everyone means everyone. This applies to all of us, including me and the executive team. …
The corporate memo is both carrot “AI is great” and stick “You will get judged on AI use.” AI is so helpful for productivity that all must take advantage of AI.
It’s a challenge to any employee to determine how and where it can help them. He challenges his employees to both use AI and share learning, so the company as a whole learns best practices for leveraging AI.
Mandates on AI influencing performance or additional headcount can be problematic. Having to ask the question “Can AI do this task?” before you get more headcount and putting AI use into reviews could turn AI from a productivity ally into a competitor. It could add to employee stress about AI if learning around best practices isn’t in place. It could backfire fully if AI doesn’t deliver hoped-for productivity gains.
The Core of Staying Current is Using AI
Part of his message about AI is one we can all take to heart:
What we have learned so far is that using AI well is a skill that needs to be carefully learned by… using it a lot. It’s just too unlike everything else.
You can apply parts of the CEO’s message to his employees as best practices for your own AI use and learning:
Use AI a lot. If you want to learn a skill, you need to become a practitioner. The best way to get the most out of AI is to get good at using AI; the best way to learn how to use AI well is to use AI a lot. AI is evolving, so what was a best practice in 2023 may be less applicable now, e.g., prompt engineering has changed.
Use AI in drafting and prototyping. Whether it’s coding, drafting reports, wrestling data, or whipping up designs and interfaces, AI can accelerate creating artifacts of knowledge. AI is a great tool for researching and prototyping more quickly. Even if AI isn’t perfect and you are doing final polishing, writing, or coding, it saves much time going from zero to completion.
Conclusion - AI and Jobs
AI will totally change Shopify, our work, and the rest of our lives. We're all in on this! I couldn't think of a better place to be part of this truly unprecedented change than being here. You don't just get a front-row seat but are surrounded by a whole company learning and pushing things forward together. – Shopify CEO Tobi Lutke
One trailblazing CEO’s memo doesn’t mean the entire world will be on AI soon, but AI adoption uptake has been phenomenal. In the AI era, most jobs will be AI-related, because people in all fields will use AI to be more productive.
Some fields will adopt sooner than others. In February, Anthropic published the Anthropic Economic Index, a study of how AI, more specifically their Claude AI model, is being used in the workplace. They found the largest uptake of AI in computer and math-related fields, as well as uptake in the arts, education, engineering, and life sciences.
Anthropic also found that while only a few people are using AI extensively, many were using AI moderately, specifically finding that in 36% of jobs, workers were using AI for at least 25% of their tasks.
Those worried about AI taking their jobs can take some comfort in that AI probably can’t do most of their job, but AI can probably speed up some tasks on their job. That’s a win/win; keep your job but get more productive. However, AI is improving so rapidly that the limits on what AI cannot do keep shrinking by the day.
PostScript: How to Use LLMs & Accelerate Your AI Learning
If you want to learn more about using AI, there’s our Substack and many other resources to help get you informed. For example, AI pioneer Andrej Karpathy has a video on how to use LLMs.
If you don’t have time to watch a two-hour video, here’s an AI prompt I use to accelerate my learning. I use Gemini’s 2.0 Flash and ask for a YouTube summary, so I can skim YouTube content and weed out fluff. It’s a simple prompt:
Summarize this YouTube video https://www.youtube. com/watch?v=EWvNQjAaOHw
A simple prompt with the YouTube link and Gemini 2.0 Flash reads in and quickly summarizes the video in Cliff Notes style. Here’s the output for the Karpathy video about LLMs:
Here's a summary of the key points discussed in the video:
Introduction to LLMs and the Ecosystem: The video introduces LLMs, with ChatGPT as the first widely used text-based LLM. It mentions the expanding ecosystem with alternatives like Google's Gemini and Anthropic's Claude, and resources for tracking model performance.
Understanding LLM Interaction: The interaction with LLMs involves text input and output. The video explains the concepts of tokens, context windows, and how LLMs process information. It also touches upon the pre-training and post-training aspects of LLM development.
Practical Applications and Considerations: LLMs can answer knowledge-based questions but are limited by their training data's cutoff date. The video advises starting new chats for different topics and being mindful of the capabilities and costs of different models.
Advanced LLM Capabilities: The video explores advanced features like thinking models, tool use (internet search), deep research, file uploads, Python interpreter, advanced data analysis, cloud artifacts, and code generation tools.
Multimodal Interaction: The video discusses how LLMs are evolving to interact with audio (speech-to-text, text-to-speech, voice imitation), video (analyzing video input), and images (image analysis and generation).
Quality of Life Features: Features like memories of past conversations, custom instructions, and custom GPTs for specific tasks are highlighted.
Conclusion: The LLM landscape is rapidly evolving, and users should consider several factors when choosing and using LLMs. While ChatGPT is powerful, other LLMs offer unique strengths.
Aside from summarizing content, another way to accelerate learning is generate specific report with Deep Research. I even used this to help me fix a settings problem on my Samsung phone. Deep Research gives you a customized tutorial, usually it is much better than earlier methods of “use search engine for 10 minutes until you find the answer.” Google has audio outputs now as well from their Deep Research, so you can create a custom report and turn it into a podcast.
Learning is becoming much easier with AI. It’s a great was to stay ahead on AI with AI’s help.