The AI Daily Brief: Artificial Intelligence News and Analysis - Shopify's AI Memo Shows the Future of AI at Work

Episode Date: April 9, 2025

Shopify CEO Tobi Lutke generated some serious buzz with a leaked internal memo where he laid out 6 mandates regarding "reflexive AI usage" inside the company, including the media attention capturing pronouncement that no teams would be able to hire new employees until they had demonstrated that AI couldn't do the job. Plus, Cursor hits 1m users.Get Ad Free AI Daily Brief: ⁠⁠⁠https://patreon.com/AIDailyBrief⁠⁠⁠Brought to you by:KPMG – Go to ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://kpmg.com/ai⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠ to learn more about how KPMG can help you drive value with our AI solutions.Vanta - Simplify compliance - ⁠⁠⁠⁠⁠⁠⁠https://vanta.com/nlwPlumb - The Automation Platform for AI Experts - https://useplumb.com/nlwThe Agent Readiness Audit from Superintelligent - Go to https://besuper.ai/ to request your company's agent readiness score.The AI Daily Brief helps you understand the most important news and discussions in AI. Subscribe to the podcast version of The AI Daily Brief wherever you listen: https://pod.link/1680633614Subscribe to the newsletter: https://aidailybrief.beehiiv.com/Join our Discord: https://bit.ly/aibreakdown

Today on the AI Daily Brief, a leaked internal memo from Shopify that shows the future of AI at the workplace. Before that on the headlines, Cursor hits a million users. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI. To join the conversation, follow the Discord link in our show notes. Welcome back to the AI Daily Brief Headlines edition, all the daily AI news you need in around five minutes. We kick off today with a story around the popularity of AI coding platform Cursor, which has now officially hit a million users honestly, without really trying. Platform creator Enesphere has managed to catch lightning in a bottle

with a three-year-old company not spending a dollar on marketing and yet still becoming one of the fastest growing startups ever. In January, Enesphere reported a $100 million in annual recurring revenue and according to Bloomberg sources, that figure had doubled by March. The platform is now seeing a million daily active users. Cursor currently offers two subscription tiers, a $20 per month account for individuals and a $40 business account. AnySphere president, Oscar Shulls mentioned that nearly all of Cursor's revenue still comes from individual users, many of whom work at companies that aren't paying for the tool directly. This suggests, of course, that adoption is bottom up, with programmers using Cursor on their own initiative because they want

to improve their productivity. This has certainly been our experience and how we see it happening, even if people are using Cursor for work, they're doing it outside on their own time or on their own dime. At the same time, this year, Schultz says that the company will begin actively courting enterprise clients. They currently have 14,000 businesses signed up, despite a deliberately obscure onboarding process. The company now has hired their first salespeople and say they had over 4,000 companies looking to test the product in February alone. AnySphere is currently in talks to raise at a $10 billion valuation,

which obviously would help them expand from their just 60-person company right now. Next up, a little interesting nugget in the OpenAI world. The company is considering acquiring Sam Altman and iPhone designer Johnny Ives' AI device startup. The information reports that OpenAI executives have discussed a $500 million acquisition price, but that a partnership was also explored. Their sources say that right now the company has only created preliminary designs rather than full prototypes. Potential designs include a phone without a screen and other modalities for household devices. The acquisition would be a very expensive aqua hire and would include the team of

engineers currently working on the device. Reports suggested that Altman is working closely with the startup but isn't a founder and has an unknown stake in the company. I think what gets people excited about this is just the idea that the guy who designed the iPhone might actually be fully unleashed to think about the new user experience of an AI-powered world. At the same time, it does feel a little early for this sort of acquisition, but who knows, OpenAI has a fat new valuation of $300 billion and money to spend. Now, speaking of talent and how much it costs to play in the AI game, the battle for AI has now grown so intense that Google would rather pay staff to do nothing rather than see them jump shipped to a competitor. According to Business Insider, Google DeepMind has

allegedly resorted to using aggressive non-compete agreements. Some staff are prevented from working for a for up to a year. The report states that staff who want to leave have been put on extended garden leave, getting paid to wait out the duration of their non-competes. This practice is very foreign to the tech industry, with California's ban on non-compete agreements dating way, way back. The Federal Trade Commission recently banned non-competes nationwide, and as of last year, California is attempting to invalidate foreign non-competes for residents. Internationally, sectors like finance make extensive use of non-competes to protect proprietary information and prevent the loss of human capital. DeepMind has been able to make use of non-competes as they're based in London,

where the practice is still legal. Commentsing on the reporting, Google said, our employment contracts are in line with market standards. Given the sensitive nature of our work, we use non-compete selectively to protect our legitimate interests. Former staffers say that six-month non-competes are common for regular DeepMind employees, while some senior researchers have year-long agreements. This is not all that surprising, given that large amounts of churn and top AI talent has been a hallmark of the AI industry. DeepMind itself suffered a ton of brain drain in prior years, with co-founder Mustafa Sullyman leaving in 2022, and Mistral co-founder Arthur Mensch departing in 2023. At the same time, Google has also participated in their fair share of talent

poaching, aqua hiring the character AI team, including co-founder Noam Shazir last year. For DeepMind staff looking to jump ship, these non-competes are obviously a massive problem. One former employee told Business Insider, who wants to sign you for starting in a year? That's forever in AI. The issue is especially bad for staff who want to move to an exciting startup, with few early-stage companies able to even wait six months to hire. The former employee said, AI is interesting. It seems like the first time in my career that you have this insane race like a space race. People really feel like to be six months ahead, a year ahead, could make all the difference.

Lastly today, an update in this whole meta situation from yesterday, the company has denied rumors that the new Lama4 models were specifically tuned to hit benchmarks. Metas VP of Generative AI, Ahmed El Dali, posted on X, were hearing some reports of mixed quality across different services. Since we dropped the models as soon as they were ready, we expect it'll take several several days for all of the public implementations to get dialed in. We'll keep working through our bug fixes and onboarding partners. We've also heard claims that we trained on test sets. That's simply not true, and we would never do that. Our best understanding is that the variable

quality people are seeing is due to needing to stabilize implementations. Now, remember, these rumors originated from a Reddit post translating a post from Chinese social media. Someone claiming they were a meta-engineer said that they had resigned over pressure to game the benchmarks. They said that company leadership suggested blending test sets from various benchmarks during the post-training process, aiming to meet the targets across various metrics and produce a presentable result. Yesterday brought a whole new wave of controversy as the AI community dug into LAMIF's performance. One of META's key claims was ranking second on LM Arena, a subjective benchmark where users vote for which model produces the most pleasing output.

LM Arena has now released all 2,000 head-to-head battles involving Lama4 for public review. They noted that, quote, meta's interpretation of our policy did not match what we expected from model providers. Meta should have made it clearer that Lama4 Maverick O326 Experimental was a customized model to optimize for human preference. In essence, Alamarina is calling out meta for deploying a fine-tune instead of a base model for benchmarking, although they did also note that the practice wasn't technically against the rules. AI developer Vic Kaye dug through the results giving one example where Lama 4 defeated Claude 3.5

sonnet. He commented, this is the clearest evidence that no one should take these rankings seriously. In this example, it's super yappy and factually inaccurate, and yet the user votes for Lama 4. The rest aren't any better. Many in the comments question whether real users were even voting on these outputs or if LAM Arena is infested with bots. Ultimately, I think the big takeaway, and really the only one that makes any sense, is to just put less emphasis on the benchmarks. But then again, how are we going to rate new things? I don't know. I don't have the answers. For now, that's going to do it for today's AID Daily Brief Headlines edition. Next up the main

episode. A quick note before we get into today's ads, for those of you who are looking for an ad-free experience, we are now up on Patreon. You can go to patreon.com slash AI Daily Brief. Right now, the benefits of this are that you will get the episodes without ads, and they will also come out a little bit earlier. We'll be exploring things like community and additional content in the weeks to come, but for now, I've heard from lots of you that you are looking for an ad-free experience, and so again, it's Patreon.com slash AI Daily Brief. Thanks as always for supporting the show. Today's episode is brought to you by Vanta. Trust isn't just earned, it's demanded. Whether you're a startup founder, navigator,

your first audit or a seasoned security professional scaling your GRC program, proving your commitment to security has never been more critical or more complex. That's where Vanta comes in. Businesses use Vanta to establish trust by automating compliance needs across over 35 frameworks like SOC2 and ISO-27001. Centralized security workflows, complete questionnaires up to 5X faster, and proactively manage vendor risk. Vanta can help you start or scale up your security program by connecting you with auditors and experts to conduct your audit and set up your security program quickly. Plus, with automation and AI throughout the platform, Vanta gives you time back,

so you can focus on building your company. Join over 9,000 global companies like Atlassian, Quora, and Factory, who use Vantage to manage risk and prove security in real time. For a limited time, this audience gets $1,000 off Vanta at vanta.com slash NLW. That's V-A-N-T-A-D-W for $1,000 off. Hey listeners, are you tasked with the safe deployment and use of trustworthy AI? KPMG has a first of its kind AI Risk and Controls Guide, which provides a structured approach for organizations to begin identifying AI risks

and design controls to mitigate threats. What makes KPMG's AI Risks and Controls Guide different is that it outlines practical control considerations to help businesses manage risks and accelerate value. To learn more, go to www.kpmG.us. guide. That's www.kmg.us. slash AI guide. Today's episode is brought to you by Plum. If you're building agentic workflows for clients or colleagues, it's time to take another

look at Plum. Plum is where AI experts create, deploy, manage, and monetize complex automations. With features like one-click updates that reach all your subscribers, user-level variables for personalization, and the ability to protect your prompts and workflow IP, it's the best place to grow your AI automation practice. serve twice the clients in half the time with plum. Sign up today at useplum.com. That's U.S.E.PLUMB.com forward slash NLW. Welcome back to the AI Daily Brief. If you were anywhere around either LinkedIn or X over the last 24 hours or so, you probably have already seen, or at least

seen conversation about this internal memo from Shopify CEO Tobias Lutkey. Now, this was apparently in the process of being leaked when Toby decided to just post the whole thing so that it could be misappropriated or miscontextualized. What's capturing people's attention with this is that this is not a gentle statement of suggestion when it comes to AI. This is a new mandate from on high around how Shopify is going to move to a reflexive AI posture with everyone using AI in a fluent and deeply integrated way. Capturing the most attention, of course, is the provision that says that Shopify folks are not allowed to request new hires until they have confirmed that AI can't do that job, as this of course cuts right into fears around AI-related job displacement. But the story is much bigger than that.

This is, I think, a preview of the future and a document that while seeming perhaps on the vanguard or even to some extreme today will seem within a very short period of time, probably about a year, completely obvious. So let's actually read it first, and then we'll dig into it. It's not all that long. When he reposted it to Twitter, Toby titled it, reflexive AI usage is now a baseline expectation at Shopify. He writes, Team, we are entering a time where more merchants and entrepreneurs could be created than any other in history. We often talk about bringing down the complexity curve to allow more people to choose this as a career. Each step along the entrepreneurial path is rife with decisions requiring skill, judgment, and knowledge.

Having AI alongside the journey and increasingly doing not just the consultation, but also doing the work for our merchants is a mind-blowing step function change here. Our task 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. Maybe you are already there and find this memo puzzling.

In that case, you already use AI as a thought partner, deep researcher, critic, tutor, or pair programmer. I use it all the time, but even I feel I'm only scratching the surface. It's the most rapid shift to how work is done that I've seen in my career, and I've been pretty clear about my enthusiasm for it. You've heard me talk about AI and weekly videos, podcast, town halls, and summit. Last summer, I used agents to create my talk and presented about that. I did this as a call to action and invitation for everyone to tinker with AI, to dispel any skepticism or confusion that this matters at all levels. Many of you took up that call,

and all of us who did have been an absolute awe of the new capabilities and tools that AI can deliver to augment our skills, crafts, and fill in our gaps. 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. The call to tinker with it was the right one, but it was too much of a suggestion. This is what I want to change here today. We have also learned that as opposed to most tools, AI acts as a multiplier. We are all lucky to work with some amazing colleagues, the kind who contribute 10x of what was previously thought possible. It's my favorite thing about this company. And what's even more amazing is that, for the first time, we see the tools become 10x themselves.

I've seen many of these people approach implausible tasks, ones we wouldn't have even chosen to tackle before, with reflexive and brilliant usage of AI to get 100x the work done. In my On Leadership memo years ago, I described Shopify as a Red Queen race based on the Alice and Wonderland story. You have to keep running just to stay still. In a company growing 20 to 40% every year, you must improve by at least that every year just to re-qualify. This goes for me as well as everyone else. This sounds daunting, but given the nature of the tools, this doesn't even sound terribly ambitious to me anymore. It's also exactly the kind of environment that our top performers tell us they want. Learning together surrounded by people who are also on their own journey of

personal growth, and working on worthwhile, meaningful, and hard problems is precisely the environment Shopify was created to provide. This represents both an opportunity and a requirement, deeply connected to our core values of be a constant learner and thrive on change. These aren't just aspirational phrases. They're fundamental expectations that come with being a part of this world-class team. This is what we founders wanted, and this is what we built. And so here we get to the meat of it. What this means. One, 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. Frankly, I don't think it's feasible to opt out of learning the skill of applying AI in your craft.

You're welcome to try, but I want to be honest, I cannot see this working out today and definitely not tomorrow. Stagnation is almost certain and stagnation is slow motion failure. If you're not climbing, you're sliding. Two, AI must be a part of your GSD prototype phase. The prototype phase of any GSD project should be dominated by AI exploration. prototypes are meant for learning and creating information. AI dramatically accelerates this process. You can learn to produce something other teams can look at, use, and reason about in a fraction of the time it used to take. Three, we will add AI usage questions to our performance and peer review

questionnaire. Learning to use AI well is an unobvious skill. My sense is that a lot of people give up after writing a prompt and not getting the ideal thing back immediately. Learning to prompt and load context is important and getting peers to provide feedback on how this is going will be valuable. Four, learning is self-directed but share what you learned. You have actually to as much of the cutting-edge AI tools as possible. There is chat.com Shopify.com, which we've had for years now. Developers have proxy, copilot, cursor, Claudecode, all pre-tooled and ready to go. We'll learn and adapt together as a team. We'll be sharing wins and losses with each other as we experiment with new AI capabilities, and we'll dedicate time to AI

integration in our monthly business reviews and product development cycles. Slack and Vault have lots of places where people share prompts that they developed. 5. 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? This question can lead to really fun discussions in projects. Six, everyone means everyone. This applies to all of us, including me and the executive team. Finally, he concludes the path forward. 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. Our job is to figure out what entrepreneurship looks like in a world where AI is universally available, and I intend for us to do the best possible job of that, and to do that, I need everyone's help. I already laid out a lot of the AI projects and the themes this year. Our roadmap is clear, and our product will better match our mission. What we need to succeed is our collective sum total skill and ambition at applying our craft multiplied by AI for the benefit of our merchants. All right, so that is the piece.

Like I said, a lot of the attention around it has to do with that fifth bullet about hiring. The Wall Street Journal piece is titled Shopify says no new hires unless AI can't do the job. TechCrunch's story is called Shopify CEO tells teams to consider using AI before growing headcount. And before we get into more substantive conversation, we should note that some people thought that this actually had to do with what was going on in the macro environment. Apple developer Nathan Lawrence, for example, tweeted, hard not to see this announcement as a soft hiring freeze caused by the ongoing economic situation, which is probably hitting many Shopify customers very hard.

Others chimed in that, yes, maybe this was a way to basically freeze hiring without spooking Wall Street. And while I don't think that there's nothing to this, obviously every company is trying to adapt to a very complicated and very tricky environment that doesn't have a lot of answers or clarity around how the future is going to evolve. It strikes me that to the extent that that is true, its main force is accelerating something which is clearly already a priority for this company. And so at least for our purposes, I think we don't have to dwell too much on how much that macro environment is playing into this. particular policy right now. To the extent that there was concern or critique about this, it usually fell into the category of either one. This isn't the right approach to executive leadership. Basically forcing things by dictate isn't likely to work. I will share in a minute why I think that's exactly opposite to true. But the other side is, of course, just general concern around

what the implications are for labor. And this wasn't even necessarily coming from people who thought that Shopify was doing anything wrong, but more of just a general concern. Malcolm Peralti writes, the recession we are already in and that will worsen in large part thanks the trade tariffs will massively increase the adoption of AI and companies. The latest announcement from Shopify's CEO about AI and headcount is just the start. Companies will aggressively downsize as an understanding of how to use AI matures. No position is safe, no level of talent is safe. The best thing you can do is diversify your skill set in hopes it keeps you employable. Now, I have a slightly different read on this. First of all, I think it's important here to reintroduce this framework that I have of opportunity

AI versus Efficiency AI. Efficiency AI is a way of looking at artificial intelligence, of course, as a tool for efficiency, a way to get the same outputs with fewer inputs. I've said before frequently that I think that inevitably some number of companies will view AI strictly in this way. They will view it as an exciting way to reduce their costs, maybe get some plaudits from Wall Street in the short term for doing so, but ultimately be wildly outcompeted by the companies who instead view AI as fundamentally opportunity generating. Putting opportunity AI in terms of labor, those who see AI as a way to radically expand the capability of their teams

to deliver new products, new services, better products, better services, and whose focus instead of on cost savings on expansion and growth. Ultimately, the companies that win are not the companies that save the most. They're the companies that grow the most. And I think that even to the extent that part of the immediate term motivation for Shopify is bracing for recession, which could, more significantly impact them given the small business entrepreneurs that they deal with and who are their major customers. They're still very clearly coming at this from that

opportunity AI, growth AI type of mindset. I mean, Toby articulates extremely clearly here that the company has to grow 20 to 40% every year just to stay still. And that because of that, it's people have to get 20 to 40% better year over year just to stay still. Thinking in that way and trying to exceed the stasis that comes from base level growth is definitionally an opportunity AI mindset. And I think it's really important to note that number five, the mandate to prove that AI can't get done what you want to get done before asking for more headcount

does not preclude more headcount. It just says you have to start by seeing how far AI can get you. To me, this is as much about a mindset shift around how people are thinking about growth as it is any sort of soft hiring freeze. Let's put it this way. This is exactly the same thing that I expect inside super-intelligent. We stretch ourselves as far as we possibly can,

every single person in this company, with AI, before we consider hiring more people. And yes, of course, in our startup world, that's because of resource constraints, but with big enough ambition, everyone is resource-constrained. If you're trying to double your revenue as a big public company, even if you have all of the resources in the world theoretically

compared to a small startup, you are still definitionally resource constrained. And so I think that this approach to hiring is just going to seem like common sense in the future. Now, that's not to say that there aren't some challenging implications here. In fact, I think it's very likely that this sort of becomes the default type of, quote-unquote, job displacement that we see in the short term from AI. It's going to be about hiring freezes and hiring slowdowns and people no longer hiring junior versions of themselves

than it is about cutting out people who currently exist and are employed. Because even if you are in that growth mindset, there's going to be a big transitional period where you really want to see how far your existing team can get superpowered by these AI tools before you start throwing more people at it. Now, one unanswered question that this brings up is absolutely how we bring new people into the workforce, what the future of mentorship looks like. If no senior people who do X ever hire junior people who do X because they're just hiring agents who do X, how are we going to make that transition? Now, I actually have a ton of ideas. I'm fairly optimistic here, but I think it's important to recognize that there is theoretically a challenge there that we will

want to address. In any case, this announcement was about a lot more than headcount. In fact, the thing that stood out most to me was the mandate of it. And this is, I think, the biggest shift that you're going to see. For a couple of years, there has been a strong encouragement from many organizations to start using and experimenting with AI. At the same time, there have been a couple of things holding organizations back. One of them is limiting the tools that are available to their people for concerns around security, privacy, etc. Another is sort of an illegitimacy of leadership. Basically, if employees don't see their managers and executives actually making these transitions, they're less likely to as well. A third piece of it is just the normal inertia.

Suggestions are only going to take you so far. And that's why it both doesn't surprise me, that we are now seeing the first mandates, and why I think this is exactly what's needed. Ultimately, people inside big companies are going to adapt to whatever they are evaluated on, and that's why this third bullet is, in some ways, I think, the most significant of these. We will add AI usage questions to our performance and peer review questionnaire. In other words, you will be judged on whether you're actually using AI or not. I don't think anything short of that could actually prompt the change that's going to be required. And of course, bullet number six that this applies to everyone is also related to this same

sort of leadership dynamic of this. Ultimately, what matters is that there is going to be accountability for this AI transformation. Now, one thing that I think that Shopify is doing well here that other organizations could take a hint from is this idea of learning being self-directed and being more open and flexible to the tools that are available to people. Like I said, another big constraint on adoption is, of course, only having some crappy old version of co-pilot that's totally underpowered relative to all the things you're using as an individual. It's why we have so many people signing in with their g-mails and using state-of-the-art tools instead of bumbling around with whatever their work said they could use.

One of the single best things that big organizations could do is relax their defensive postures when it comes to experimenting with new tools, because otherwise these smaller and mid-sized organizations are just absolutely going to eat them for lunch. I did also want to mention quickly this idea of AI being a part of any sort of prototype or planning phase. this is maybe the most specific injunction here, and also one that we're seeing just everywhere now. AI in general creates a hugely differentiated opportunity to test out more ideas and whittle them down, rather than trying to figure out the best one before you've actually run any tests.

What's more, the rise of these vibe coding tools has totally changed who has access to build and prototype things. Already we're seeing inside our own company and in other startups that we know, basically a ban on feature discussions that don't involve actual working prototypes because there's just no reason not to be able to do that now. And in fact, going back to the point just before about enterprises relaxing their defensive posture, I actually think that this is one area where even very defensive organizations can move the quickest. Sure, maybe lovable and bolt and these tools are not ready for enterprise prime time when it comes to actually integrating with big morass-like code bases that are touched by thousands of people. But when it comes to mocking up new features, where you know that that mock-up is, is never meant to see the light of day from an actual code perspective. It's just meant to

feed the speed with which you can make new suggestions and have new conversations about new opportunities. It actually seems like something that perfectly hits that low stakes but high impact Goldilocks spot. I will be eagerly awaiting the first of you corporate executives to email me and tell me that your teams have embraced the vibe prototyping life. So as I said at the start, I believe that these Toby rules, or whatever they end up being called, really do reflect where the world is headed. I would encourage that to the extent that these feel dramatic now, the more dramatic they feel, probably the farther you or your organization are behind. Not every organization has to adopt these exact same six mandates, but I think that

they should be thinking in these sort of mandated type terms. And again, to reinforce, what's ultimately exciting about this is that I do believe that Shopify is coming at this with an opportunity AI mindset. And my strong, strong guess is that this makes the company stronger, better performing, and ultimately is more likely to grow headcount in the medium and long term than if they didn't have these mandates in place. Anyways, guys, that is going to do it for today's AI Daily Brief. Appreciate you listening, as always. Until next time, peace.