--- Source: https://37signals.com/podcast/ai-revisited ---

SEASON 2 - EPISODE 180 - January 21, 2026
AI Revisited
Listen now
AI has moved fast, and so has 37signals’ thinking. This week, host Kimberly Rhodes talks to co-founder and CTO David Heinemeier Hansson about the progress AI has made over the last few months. David shares what’s changed, what’s actually useful now, and why his outlook has grown more optimistic.
Watch the full video episode on YouTube
Key Takeaways
00:11
- How today’s AI feels dramatically more powerful
05:05
- David’s real-world experience working with AI agents
23:36
- ow AI can help non-programmers
31:05
- Why the upside of AI outweighs the potential risks
Links & Resources
Fizzy – a new take on kanban
O’Saasy License Agreement
Record a video question for the podcast
Books by 37signals
30-day free trial of HEY
HEY World
The REWORK Podcast
Shop the REWORK Merch Store
The 37signals Dev Blog
37signals on YouTube
37signals on X
Sign up for a 30-day free trial at Basecamp.com
Transcript
Kimberly (00:00):
Welcome to REWORK, a podcast by 37signals about the better way to work and run your business. I’m Kimberly Rhodes, joined this week by David Heinemeier Hansson, co-founder and CTO of 37signals. Well, we have gotten a lot of questions over the years about AI and incorporating AI into our products. We’ve typically said, “We’re just not quite there yet. We’re seeing how things go,” but I feel like there’s been a little bit of a shift recently, internally, at least from announcements that I’ve seen, so I thought we would talk a little bit about that today. David, why don’t you just jump right in? I know we’ve kind of not been hesitant. I don’t know if hesitant is the right word, but it seems like we’re jumping in a bit more now.
David (00:40):
Yeah. I think for me, AI really changed in the later half of 2025. The models got better as they’ve been getting better for a while, but we gave them more powers, especially when it comes to development. There’s a new, or not even that new, but approach of instead of having the AI being in your editor, auto completing things, trying to help you write something that you’re in the process of writing, they moved out of that and into the terminal running in agent mode, which basically just means that they’re running on their own from a script, from a plan where you tell them to do something and they go off using the tools available on the computer. They use your terminal, they run bash commands, they might even run some programming. They start looking things up on the web. They’ve been doing that for a while. But when you combine these things that they have control over your computer, which sounds a little scary, but when you see it in action as an agent doing development alongside you, not sort of on top of you, it really changes the game, not just in how it feels, which to me was a really important part of it.
(01:55):
I played around a fair bit with the auto completion paradigm of AI as it was for developers in the initial phase, and I didn’t like it at all. I did not like this idea that I have a thought in my head coming out through my keys and then boom, the AI tries to … Did you think of this? Did you think of that? Can you shut up for a second and just let me type the damn thought out of my head? Now, I know there are others who rather enjoyed that and saw some great benefit out of it. I didn’t like it. And that was where a lot of my opposition to using AI as a developer tool was coming from, that I could see the advantages, I could see the promise, at least I should say, but it didn’t feel like it was in a format that was compatible with how I wanted to work.
(02:42):
I don’t like working in IDEs, these heavy duty development machines. I like a text editor. I like to type my little code out by hand. And it felt like it was interfering with that. It was stepping over my toes. But this new way where they run in what’s being called agent mode, where they’re on their own doing stuff that you set them off to do sometimes for 10 seconds, 30 seconds, other times for multiple minutes really changes things. So you have this duality here where the mode of using them for me as a developer totally changed. It went from auto complete to agentic. I mean, I’m having a hard time even saying that word agentic because it’s so damn buzzwordy at the moment that you almost be like, all right, make sure you don’t say some bullshit here because agentic sounds a little bit like bullshit.
(03:35):
But I used the term anyway. There’s agentic mode where they’re doing work on their own, that was huge. And then the step up in capacity is really huge too. The initial phase where I tried to use AI, even in early ’25, I kept checking in every month through every new model drop. I’m like, all right, cool. Everyone’s getting excited about this. Let me try it for something real. Let me give it a real code base, a real problem, see if it can help me. And I kept seeing like, do you know what? There’s something here, but it’s not there yet. It’s not actually helping me. I can see the promise. I can be amazed that we’re even making computers do this, even if I don’t actually want to keep any of the work it’s doing. And again, here in the second half of 2025, now coming into ’26, it really just changed.
(04:24):
The models with their mode of working became so good that when I asked them to help me, they were actually helpful in a way where I wanted to keep the majority of what they produced. And I wrote about this earlier this week about promoting AI agents and sort of like a double meaning that both, hey, you should check this out. It’s actually quite cool what’s been going on. So if you’ve been, I don’t know, using ChatGPT or using Cursor or any of the auto completion mode, try out this agentic thing if you’re working as a software developer or a designer working directly under code, you should check it out because it’s really cool. But then for me personally, there’s a promotion here where the AI that I’ve been using for the past several years just went from a text input box where I would ask at things, help me with APIs, maybe even double check something if it was uncertain or wanted the second opinion to actually doing work where the product of that work could be kept.
(05:28):
I’d still tweak it. I’d amend it. Sometimes I’d just chuck it out. It wasn’t good enough, but I’ve been shocked how often it’s like 80% there. And sometimes I could get the last 20% by just iterating with the agent. And more often I’d say, I just jump in and do the last 20%. If someone can take a problem that I’m facing and I want done and do 80% of it, that’s amazing. If you could even just do 20% of it and not fill the rest of it with bullshit that would take me even longer to pull out, that’s still amazing. And it’s that personal experience that these agents, this AI that we have access to now have really just taken a leap that has made me question my own relationship with the AI, how I feel about it, how I’m involved I want it to be in my own work.
(06:21):
And then what are we doing at 37signals? This party also came from Jeremy who has been pioneering ways of putting these agents to work on the toil that we have. And we have a fair bit of toil, programmer work that isn’t the most exciting thing in the world, but it’s very important and needs to be done. Things like reviewing security reports. So we use a system called HackerOne where external researchers, and I use that term very, very broadly because researcher is a really big word to use about 90% of the reports we get that are more like bullshit. But nonetheless, when you find or someone finds an actual security research or find something important, it’s obviously a huge value because if they find it before a hacker does, that’s great for us. And that’s why we participate in the HackerOne program where we have these bounties, that’s what motivates these researchers to submit their findings to us.
(07:25):
I think their payouts of up to $5,000, or maybe it’s even more at this point if you find something truly catastrophic that could give you access to execution context or exfiltrate data or whatever have you. The vast, vast, vast majority of reports we get, even the good ones on that, they’re like, oh yeah, I guess if you really string all this stuff together, you could cause some harm to our system in some way. But as I said, 90% of it is not of high quality, but it takes us quite a long time to review it because we got to check. You don’t know in advance what’s the great report where you’re going to find something really important that needs to be fixed and what’s kind of just low effort sludge, as I called it, created by humans or maybe by not so good AI submitted to us.
(08:14):
So Jeremy came up with a system where we can use AI to give it access to the history of the reports we’ve gotten, give it access to sort of a scoring of the people submitting reports. Has this person been submitting nonsense in the past? Then maybe take that into consideration when you’re weighing whether this report merits more delicate inquiry. And it’s been a huge hit. It’s been a huge step forward for us that we can save a lot of time by having the toil of reviewing these reports pre-cooked by an AI agent that can go through it and then we can spend more of our time on the high value stuff and less time on the low value stuff, which in some ways is kind of just taking the concept of spam detection and applying it to a slightly different context. This isn’t spam, this is just low signal and AI turns out to be really good at that, right?
(09:10):
It’s really good at filtering noise from signal. So that’s one example of something that has really helped us. And then we’ve started looking into where else can it take some of this toilet away. One thing we do, for example, is we do these things called console reviews where we, on a biweekly basis, review all the access logs when programmers have used access to our production systems and our production data in service of doing investigations or inquiries into customer reports. And we need to make sure that no one we have on staff is accessing data in any way other than what they have been granted permission to do by customers. So we review the logs and we make sure that people have got the right permissions and all of that stuff. And it’s kind of tedious. And it’s the same sort of thing where I don’t know if we’ve … oh, that’s not true.
(10:02):
When we first started that program, we found some bad habits, some sort of sloppy ways of getting access to too much data at once and we shaped those things up. But the last many, many reports that I’ve read from when we do this by hand have said, all right, here’s 42 accesses to personal information that we have in these systems. They were all vetted, they were all good, they were all within jurisdiction. Now, that’s the kind of work that kind of does get a little cumbersome or boring, even at times when you’ve been doing the same thing for a long time and you keep finding nothing. Well, do you know what? AI is perfectly patient to do the same work over and over and over again, and you can train it quite well to look at some of these things and see, find the operation, find something that’s not quite right.
(10:49):
So that’s another part of it. And then finally, we’ve also had some good success assisting us on on-call or performance issues where we have some sort of degradation. We need to find out what’s going on here. Can you look at the logs? Can you look at our exception systems, our performance monitoring systems? Tell me what’s going on. And it’s shocking at times that you give the AI access to this system and it’ll pull all or stuff and do all sorts of queries and it comes up with a goddamn answer where you just have to sit back and go, that’s pretty impressive. No, that’s damn impressive. And how did it do it so quickly? So these are some of those areas, toil, investigations, inquiries that require pulling a bunch of things together where we’re finding that AI is starting to live up to just a fraction of the hype and the promise that we’ve been pumped full of for the last, what, two and a half, three years, right?
(11:46):
It’s finally starting to get to the point where this isn’t just, oh, that’s cool to, oh, that’s really useful. So both of those two experiences as a programmer using AI to assist our work and as us as an organization employing these AI agents to help with some of this toil, we’re finding a new enthusiasm for applying it. Now, it’s interesting because both of these cases are actually internally facing. They’re not about features. Some of the early exploration with AI that we didn’t end up shipping was, how can we jam this into the product? And we tried certain things around search. We tried certain things about commands and auto completion, and we never really ended up with something that felt like slam dunk. This is just amazing. Everyone is going to go, wow, I’m so glad you shipped this. A lot of it did have this tinge of do you know what?
(12:45):
If we launched this, people are going to say it’s a little bit of AI slob. It’s about putting on a sticker of something where it’s not uniformly better after we’ve introduced this, and that’s not what we want to launch. So in that sense, I don’t mind being a little later to the game. And then when we do show up, we ship something that’s useful. Now, all of this work with our own personal toil reduction, our programmer systems, our design systems that we’re getting out of our engagement with AI, as it has happened since probably even just the last three, four months of 2025, has given us a newfound seal to try to find some of the areas where we can deliver that kind of value, that kind of unequivocally positive impact in the product itself. And we have a bunch of really neat ideas and I’m really excited about one of them specifically around Basecamp, but I don’t want to sort of just…
Kimberly (13:42):
No spoilers. …
David (13:43):
Throw open the kimono just yet because we may not launch it either, right? This is the thing with AI. You can get excited about seeing these glimmers, the shimmering of, holy crap, we’re about to enter into a new world. And you get these anecdotes, you get these experiences, these peak experiences. I had one the other day where there was a bug in the latest version of Rails and it was producing this really weird artifact when you were opening a Rails console and I just sighed because I was like, do you know what? I don’t really want to do the deep dive right now to figure out what the problem is. And I gave it to, I think it was Opus, which is Claude’s latest model, which is one of those frontier models that really does feel like it’s a serious step forward, the Opus 4.5.
(14:30):
I give it this task and the fucking thing starts pulling out a CD bugger and all sorts of tools that I know they exist, but this is not something where I just go like, all right, I got 45 seconds here. Let me just crack open this toolbox right now. And I just had to sit there as it was thinking through and I was seeing all the thinking. This is what I love about these new thinking models, you get to see how it reasons with itself whether half of it is just for our amusement and edification or whether it’s actually what it’s doing underneath, I don’t really care. The show is spectacular. It’s going through these motions like, oh, I think there might be a problem with this thing. Maybe it’s something it’s linking to. I should try to pull up the CD bugger and connect it to the process. And I’m like, wait, you can do that?
(15:21):
How do you know how to do that? And then it keeps going through and the first, I don’t know, two hypotheses as, it doesn’t pan out. It’s not what it thinks it is, but it just keeps going. And I’m just sitting there, oh wow, this is a show here. Let me see where it ends up. And it ends up just figuring it out, pinpoints exactly where the issue is, finds the commit where things went wrong, comes up with a patch that I could just put in the Basecamp code base to work around it until we fix it in the core of rails. And I simply just had to sit back and say, do you know what? I wouldn’t have done this.
(15:57):
I don’t know, maybe in theory I could have over some period of time, over some period of halfs of hours or full hours dedicated to it, but I wouldn’t have done it right now in that situation. I had other things I wanted to do. I would’ve left it on side and found a way to work around it. Now, AI is getting problems out of my way. Not everything’s like this. I’ve had other times where I’ve gotten so excited from one of those positive experiences and I ask it for something that I actually think is rather simple and it goes make a mess of things. But the ratio is changing so fast between I asked the agent something and it goes to make a mess of things and I ask the agent something and it goes to either just solve the damn problem or at least give me enough of a draft that I go like, oh, I see where you’re going here.
(16:43):
I’m going to wrap it up. Thank you very much for all the clues. This is exactly what I needed. Now I can polish it off or I can finish it. Or sometimes I even throw out its solution entirely and write it all from scratch again, but it’s given me the blueprint of where I need to go. Other times I don’t even just get one blueprint, I just get a bunch of drafts. I tried this with one job I had. I wanted to look into this MCP standard, which by the way now seems like maybe it’s kind of fading. Things do move quite fast with all this AI stuff, but I wanted to see like, all right, if I wanted to build this in a Rails kind of way, maybe we could have a new framework for it. Let me just see what a bunch of different agents can do.
(17:24):
And this is where one of my favorite new tools is a TUI, a terminal user interface called OpenCode. And OpenCode, which by the way, we just pushed into Omarchy and now it’s part of the default setup, is a way to use all of the models from all the providers at the same time, in the same interface. So you can use CloudCode, you can use Gemini, you can use OpenAI’s Codex and a bunch of other of the commercial offerings. And then you can use all these open weight models, which is kind of like open source versions of AI that are being run by commodity hardware providers. And I got five drafts from five different models on this problem that I had, and they were all quite different, but you know what? They all worked, which shocked me to no end. I gave it a fairly open-ended tasks.
(18:20):
It was not a detailed plan. I just built me this MCP connector. It’s got these three tools. This is what I want to show. Here’s how to authenticate, use a bearer token, blah, blah, blah. The agents go off and they spend about between two and a half and I think six minutes on the tasks. Each of them, every single one of them was able to use the tools available, the tests that we already had to come up with something that worked. Now in the end, I didn’t actually like any of the five enough to keep them, and I wanted to really focus on the fluidity of the interface and so forth. And they weren’t trying to do that, but I was so blown away by the fact that they all produced a working solution. Several of them had really good ideas that I hadn’t actually thought about myself.
(19:10):
And I could then take it all, put it together, and then come up with something of my own, but much faster because I could get to running software in that. I could get something that worked, hook it up to my MCP debugger and see hello world come back out. It is one of the most exciting things I’ve seen in front of my eyes that a computer has ever done for me.
Kimberly (19:35):
That’s a bold statement.
David (19:36):
It really is. And it serves to be up there. I will credit it right up there, like the peak experience of using AI in this way as a programmer and seeing it solve a hard debugging problem or produce a draft of some code that just works the first time and how things through it is up there for me with the first time I sat down in front of a Commodore 64 to play Yie Ar Kung Fu with all the kids in the neighborhood when I was five. The mind blown experience of like, here I’m sitting with a joystick for the first time at age five or six and just going like, this is amazing. And then fast forward, I think ninth grade, ’93, 9’4, it’s got to be ’94. Whenever the first version of Netscape Navigator, which wasn’t called that, what was the first thing called?
(20:28):
I forget. Andreessen’s first version of a browser. I used that in a university and amazingly they invited all these ninth grade kids in there and we got to make a webpage and we got to hit publish and it was live on the internet for the entire world to potentially see, none of them did, but just this knowledge like, wait a minute, are you telling me that I just published something that a person in South Africa or Japan or New York could see? What? That’s crazy. That’s the level we’re at or that’s the level I’m at in my excitement for AI in this moment, which is kind of wild because I think a lot of people have been blown away for a while. I’ve been blown away for the conversations I’ve been able to have with ChatGPT or other models where I just asking information or whatever, treat it as a better Google, but to see it transformed into this agent form where it’s doing work and just chugging along, asking sometimes for your feedback, other times not at all, just saying, I’m done. The problem with this is it’s so easy to fall into these hyperbolic dreams of like, holy shit, if it does this now, after we’ve only known this technology in the broader consciousness for three years, where are we next year?
(21:51):
And then people’s brains start freewheeling into like, “They’re going to change all the jobs tomorrow. They’re going to kill us all.” They’re going to do all these things. And I get that. I really do because the ramp has been so steep and the reality of where we even are today is so amazing that I forgive anyone for going, well, what is this going to be like a little further down the line, right? But I don’t even want to focus on that because nobody knows, nobody knows where this is going. Nobody knows whether LLMs are going to tap out and we’re actually almost near the top of the curve in terms of how much better they’re going to be. Nobody knows if we’re five minutes away from breakthrough that gives us AGI, artificial general intelligence. Nobody knows if it’s going to be another 10 years or it’s going to be six months.
(22:43):
So can we just focus on how good it is right now? Can we just focus on if this was all it gave us, was this capacity in this moment, how truly incredible an achievement of mankind it is to have given birth to this. And you know, as soon as you start talking about these things and these terms, you’re like, The Matrix. That’s what he said. Like he said, mankind celebrated itself as it gave birth to AI and we marveled at our ingenuity and then we set fire to the sky. So I’m trying to exist in some realm where I am endlessly fascinated and upbeat about what we’re making computers do and not falling into the trap of hyperbolic extrapolations of what the future’s going to be in either five minutes, five years or 50 years.
Kimberly (23:36):
Okay. So let me ask you this because you have mentioned a couple times having these agents work for you, but you’re always reviewing, maybe an 80%, you’re finishing the last 20% or maybe even redoing it after you’ve gotten ideas. We get a lot of questions from people who maybe are designers and are using AI to program or are beginner programmers and are using AI. How important do you think it is, like you’re obviously expert level, for someone who is not as experienced, can they get this much satisfaction out of AI as you have where you’re clearly putting in kind of that final mile?
David (24:13):
I think they can get more out of AI than I can get out of AI. All the things I’m asking AI to do, I know I could do given enough time, given enough dedication, attention, I could do it. I haven’t yet seen it do something for me where I just go like, I couldn’t have done that. No way, no how, any amount of time. That’s not what I’m experiencing. That is what non-programmers are experiencing when they’re asking the AI to build their idea and they interact with it and it produces the damn thing without them understanding a lick of what’s going on underneath. That kind of magic is closer to the magic of me playing Yie Ar Kung Fu for the first time. I didn’t know how to program a game. I was just marveling at the fact that someone did and it was this cool fighting game and the little characters were moving on the screen when I was moving the joystick and that was incredible.
(25:07):
And we’re democratizing the access to creativity in a way that very few moments in history have done so broadly. Now, I obviously look back at the history of computer programming and look at like, you know what? We’ve had other similarities here where we gave non-programmers the power to create programs. Excel is a wonderful example of that. There are companies everywhere to this day that run off monstrous Excel spreadsheet with cells doing math and all sorts of invocations. I see it all the time in racing that are essentially small programs written by non-programmers to do business tasks of various kinds that they, in many cases, either couldn’t afford, didn’t have access to, didn’t think through having a professional programmer do for them and they didn’t have the skills themselves to do it. So Excel gave them this power. Now, what’s funny is that Jason actually got started with his work in software using something sort of kind of similar, which was this program on the Mac, I believe, called FileMaker Pro.
Kimberly (26:16):
Oh, I totally used to use FileMaker Pro. Yeah.
David (26:19):
Oh, you did?
Kimberly (26:20):
Yeah, back in the day.
David (26:21):
Yes. And I’m sure you had that experience too where I’m actually creating programs. I don’t necessarily know how it’s happening, how it’s working. I know what I want and I can somehow string things together despite not having a full understanding of what that is. And I can get something that works, something that helps, something that does what I needed to do. And AI already today, not even thinking, well, can you do three months from now, six months from now? No, no. Today is doing this for tons of people and they are justifiably extremely excited. Now, that’s not to say that what these models are producing today have the level of quality where I’d say I’d want to put my personal data into it. I’ve seen enough horror examples of the AI making something that appears to work but is then leaking like a sieve or woefully insecure or whatever you have.
(27:16):
But in many cases, it also just doesn’t matter. Not everything has the criticality of a pacemaker or a bank. There are lots of kinds of programs that exist at the early stages of the criticality ladder where even if it’s catastrophically wrong about what it’s doing and it loses everything, or even deletes your computer, do you know what? That’s a risk worth taking because cost of that is not that great. And the upside of getting programs that didn’t exist yesterday because you’re letting it do this is just worth it. Again, not in all situations, not in all cases, but this is where, especially these non-programmers who’ve had this experience building something they could not have built otherwise, that’s actually kind of good, that actually looks kind of great and that was made really quickly is accelerating and they start extrapolating. Well, okay, now maybe it’s producing something that’s not super secure and whatever.
(28:14):
Well, hey, hello, innovator’s dilemma. Is that not everything? Is that not everything that starts feeling like a toy? Is that not everything where, well, it’s not good enough for my enterpris grade military demands and needs. No, maybe it’s not. Well, not maybe. It’s definitely not without supervision, but so what? That characterizes almost every single innovation we’ve had that completely upended the technology industry from day one. Everything starts out that way. Everything starts out being feeble, not enough, can’t do the top end stuff, and then it grows into it. And I think this is why this very moment is so magic because we can all see the progress. We can all see how quickly it’s improving and no one knows where it’s going to go. No one knows, are we going faster? Are we going hyperbolic now? Are we going to be five steps away from the AI writing its own AI code?
(29:13):
And then boom, that’s how you get the singularity and all the either dystopia or utopia versions of the world that people like to project. I don’t know. Holy shit, that’s exciting. To be alive right now at this moment in human history, holy crap, what a privilege. There are very few other moments where you go like, in a two-year span, how did the world potentially completely rewrite itself? Even the internet, which really has rewritten the world in almost all the ways that a computer could, took quite a lot longer than that. From when I got involved with that first HTML page in ninth grade till the internet has rewritten society for everyone broadly. That was 10 years, 12 years, maybe something like that. We’re barely three years into the public consciousness of AI existing. Now, I know there’s a long history, literally going back to the 50s of people doing research into all this stuff.
(30:17):
So it’s not like, oh my god, overnight success. How did AI just happen? Well, it was just one guy in a shed like for two weeks. No, no, no. There’s a long history of that and obviously great respect to be paid to those people, but it all entered into the public consciousness when we got ChatGPT. And already going from that, the first thing, and especially on the image generation too, I’d say this is even more obvious where you see the first image generated by Midjourney V1.
Kimberly (30:42):
With the seven fingers.
David (30:44):
Seven fingers and mutant faces. And now we’re at, yeah, there’s no way a human could tell that the latest AI models when asked to generate hyperrealistic photos taking the iPhone style isn’t actually legit or not. Boom, two and a half years, three years. So again, it’s that velocity, it’s that acceleration that’s making people so excited, making me so excited because I just love computers. I love when we make computers do new things. And I choose to be optimistic about us as a species being able to make those new capabilities do more good than harm. Because it’s really easy to imagine all the worst things that can happen.
Kimberly (31:30):
Oh yeah. There’s been many sci-fi movies telling us everything bad that’s gonna happen.
David (31:35):
Yes. Well, it’s actually, we don’t even need to use our imagination. Just watch any sci-fi movie produced in the last, I don’t know, 50 years. In fact, speaking of, I just watched 2001 of Space Odyssey, Stanley Kubrick’s movie from 1968 or 64, I think even it is. 64, I believe. And you’re like, motherfucker, how did he know? How did he know that Hal 9000 would say, “Sorry, Dave. I just can’t do that, Dave.” And now you see all these papers from the Frontier Labs going like, “Oh, when we’ve really tickled the AI, it actually tried to report us to the authorities, lock us out of our own computers and basically blackmail us into certain things.” We were like, man, that was eerily spot on from ’64. By the way, amazing movie, you should totally watch it because in some ways it is the most timely depiction of all the fears we might have about AI.
(32:35):
And on the other hand, it’s the most 64 movie you can imagine that starts with literally 15 minutes of apes running around, hitting each other over the top of the head with bones. You’re like, man, is there any content today that gets published where you just have a 15 minute scene that basically doesn’t go anywhere and have this one point at the end about an obelisk? That’s a really curious place to end up.
Kimberly (33:01):
Clearly our attention span was different in 1964 than it is currently.
David (33:06):
It was, which is why it’s a good actual challenge because as I was sitting, I was watching it on the plane. And unfortunately I say the plane also had wifi. Therefore, at any one time I had the temptation that I could hop on something more immediately simulating. And I was like, no, do you know what? In this moment, I’m going to sit down and watch a two and a half hour movie where the script could be summarized in about 90 seconds. It’s not a complicated movie. It’s an amazing movie. It’s not a complicated movie. There’s not a lot to keep track of. The plot lines, there’s about five of them. So anyway, just get to this point that we’re at this moment. Everything is amazing. Everything is scary. Everything is potential and reality at the same time. It is an incredible time to be alive.
(33:55):
And I wish for more people to take that stance in the face of uncertainty because there’s a million reasons. You could spend your whole life being worried about all sorts of things. Do you know what? Fucking in ’64 when 2001 came out, we had the threat of nuclear war. At any given time, we could all be gone and a bunch of people, fewer then than now, worried their lives sick with something that never fucking happened. So could you just wait until the bomb drops? Okay, maybe there’s a few people who need to think about it and do some scenarios and some modeling. But if you’re not one of them, and if you don’t work at Goddamn DARPA or the Pentagon or research lab, don’t spend your time worrying about the end of the world. If it’s going to come, enjoy the time you have until then.
(34:45):
Be optimistic, be happy until the final flash. When AGI shuts off the world and we have to lit the sky on fire. That’s my philosophy as a technologist too, even though I can see all the scenarios. I’m saying, do you know what? No, that’s not how I’m going to go to bed tonight. I’m going to go bed tonight going like, holy shit, this is amazing. Best time to be in love with computers and see what could happen and wake up in the morning genuinely curious of where this is going to go.
Kimberly (35:18):
I had another question, but I feel like we’re ending on such a high note. We should just wrap it up there. REWORK is a production of 37signals. You can find show notes and transcripts on our website at 37signals.com/podcast. Full video episodes are on YouTube. And if you have a question for Jason or David about a better way to work and run your business, leave us a video question. You can do that at 37signals.com/podcastquestion or send us an email to rework@37signals.com.

--- Source: https://37signals.com/podcast/ai-revisited-part-2 ---

SEASON 2 - EPISODE 186 - March 4, 2026
AI Revisited - Part 2
Listen now
Following up on an earlier conversation about AI, this episode shifts to the product side of the discussion. Host Kimberly Rhodes chats with 37signals CEO and co-founder Jason Fried about his daily AI use, what it’s helped him do more efficiently, customer expectations, and how he’s thinking about AI’s role in future product updates.
Watch the full video episode on YouTube
Key Takeaways
00:12
- Putting AI tools to work
06:42
- Using AI to reclaim time, not replace thinking
12:19
- Where 37signals sees thoughtful implementation fitting in
17:16
- Why rushing adoption can backfire
18:25
- The ongoing debate from the customer perspective
21:45
- Why workforce changes aren’t always tied to automation
25:40
- Human interaction still matters
Links & Resources
“The owner’s word weighs a ton” by Jason Fried on Signal v. Noise
Fizzy – a new take on kanban
O’Saasy License Agreement
Record a video question for the podcast
Books by 37signals
30-day free trial of HEY
HEY World
The REWORK Podcast
Shop the REWORK Merch Store
The 37signals Dev Blog
37signals on YouTube
37signals on X
Sign up for a 30-day free trial at Basecamp.com
Transcript
Kimberly (00:00):
Welcome to REWORK, a podcast by 37signals about the better way to work and run your business. I’m your host, Kimberly Rhodes, joined this week by Jason Fried, CEO of 37signals. Now, we talked a couple weeks with David about AI and this new energy around it, specifically how our programmers are using it internally. This week, I thought we would talk with Jason about his thoughts on AI and from a product perspective, where our thoughts are. So Jason, before we talk about our products, let’s maybe talk about just AI in general. Are you using this on a day-to-day basis and what kind of use are you getting out of it so far?
Jason (00:34):
Yeah. I use it for all sorts of different things. On the personal side, plenty of stuff. Business side, what I’ve been using it most for lately is actually as sort of an editor.
Kimberly (00:44):
Same.
Jason (00:44):
So I do a lot of writing. So I’m currently writing, for example, a new basecamp.com homepage. And I wrote this piece as sort of a letter form, which is kind of how I approach these things. And I wanted to make sure that I was speaking really plainly and clearly. And I thought I was, but I also wanted to really match it up with a lot of the language that our customers use very specifically. So we have this page on our site, basecamp.com/customers, which has about, I think it’s close to a thousand customer testimonials.
Kimberly (01:13):
Wow.
Jason (01:14):
And I use both Claude and ChatGPT. I kind of use them almost in competition with each other to kind of hone in on something because they each have their own style and I don’t like their house styles necessarily, but they can kind of somehow help me get somewhere that I’m comfortable with in a way that I like. So what I did was I pointed them both at the customer testimonial page and just said, internalize the language our customers are using, how they describe things, what they call things. Because for me, I might call a feature to-dos, and most people might call it tasks.
Kimberly (01:42):
Right.
Jason (01:42):
Now, I know it as to-do’s, the tool in Basecamp’s called to- dos, but maybe people are calling it tasks and maybe we should call it tasks, but I want to make sure that what I wrote will land with more people and just land in their own mental model of what they’re thinking about and how they’re thinking about the product. So I asked it to internalize all the language and then read my letter and then make some suggestions for ways to tweak the language, not to tweak the letter or the tone, but make sure that I’m more aligned with how our real world customers speak about these things and call these things and name these things. So for example, I was using things like stay on top of things, make sure you know what’s going on. And I like those phrases, but our customers just say organized.
(02:24):
They just say, “I like that I’m organized. Basecamp keeps me organized. I’m so much more organized.” So I don’t always take the suggestions from AI, but I think it’s a really good way to gut check and go, “This phrase, this word, is this lined up with what people are thinking in their head?”
(02:40):
So I’m not asking it to write the thing. I don’t like that, but I am asking it to see things that I can’t see and hold a lot of context in its head that I can’t hold and go, you know what? These four words would probably land better if they were these four words. And maybe I’ll use all four, maybe I’ll use three, whatever it is. I also tend to sometimes write in a way where I have one sentence too much in a paragraph, just like one too many sentences. And so I’ll often say, turn this into three instead of four, but don’t just concatenate them. Really, what needs to go here? And so that’s the kind of stuff I’ve been using it for primarily. That said, I also have been digging into, we’re doing Basecamp 5 right now. So I’ve been digging into some features in the product and getting in there and having it tweak some things for me and change some things for me and quickly prototype some things for me. So I could see if this idea I have in my head is even worth pursuing and it can throw some things together pretty quickly that are really, really handy that I could have done, but it would take a long time and I don’t want to sink that kind of time in to find out if something’s worth doing in the first place.
Kimberly (03:38):
Okay. When you say prototype, what do you mean? Are you having it build something for you that you can physically see?
Jason (03:45):
Yeah. So I’ll give you an example. We now have this feature in the product, but in Basecamp 5, there’s a sidebar with your pings in it. Currently in Basecamp 4, there’s a menu where you pull it down and you can see pings are like direct messages. So you can pull the menu down and see. Now we have it in the sidebar. If you want to open the sidebar, you can chat with people and not lose your place. What we do is we have these, currently we have little heads that show up, little avatars that show up when someone pings you. And if you don’t have the bar open, their head shows up with a little orange dot saying, “Hey, there’s something new from Kimberly,” or something like that. That’s what that suggests. And if you hit P in your keyboard, it opens it up and you can start chatting.
(04:18):
And that’s great, by the way, it’s great. But once I’ve clicked your avatar and I chat with you and I close the sidebar, your face is gone. And so if I want to get back to you, it’s like an extra step or two to get back to that chat. And I found like I’m often talking to the same handful of people over the course of five hours or something like that. So it’d be nice to have this menu build up of people I’ve talked to recently. We didn’t have this in the product though. We just had the new pings. And I’d asked someone to build this, but they hadn’t gotten around to it yet because they were busy with something else. So I just built it. I mean, I didn’t do the building really. Claude did the building, but I asked it to quickly prototype this idea so heads would stick around after I chatted with somebody.
(04:58):
Now in the real product, I think we have like a six-hour timeframe in which a head sticks around. If you don’t talk to that person again, it clears out. That way you’re not like having the sidebar full of people you’re not talking to, but it’s kind of kind of an active culling of people. And I was able to quickly make this work for me. It probably wasn’t like robust enough to like ship in the product, but I got it together in a way where it was working.
Kimberly (05:22):
Yeah. So you could see it.
Jason (05:24):
Yeah. I could mock it up myself, but that’s just looking at two states. I wanted to actually use it that way and see it work. And so that’s something that I would’ve had to ask someone else to do and I was able to do that myself. So I’m doing more of that. And that’s been really, really incredibly handy and like a breakthrough revelation kind of thing like, wow. And for me, it’s not so much like I can do things I couldn’t do before because there’s certain things I could do before that I just chose not to do. So it’s mostly like, this is a huge speed up of time and I don’t need to bother somebody else. And I’ve recognized how important this is because we’ve written this up in the past, like I think it’s called like the Owner’s Word Weighs a Ton or something like that, was this old post I wrote up about how when you own the place like I do and David does, and if we ask someone to do something, there’s just more weight attached to that request regardless of whether or not we’re like, “Don’t worry about it, don’t worry about it.” Somehow there’s still more weight attached to that. So people tend to sort of drop what they’re doing to help.
Kimberly (06:21):
Yeah, of course.
Jason (06:21):
I really don’t want that most of the time, but it just comes with the territory. So it’s nice to not have to ask anybody and then not pull people off of things that even if I said, “Don’t worry about it,” somehow they’d maybe get to it just because they felt obligated in some way that I don’t want to put on them, but they still do. And so I just find that I’m able to bother people less and just get some stuff in my head quicker and decide if it’s even worth pursuing or not.
Kimberly (06:42):
Okay. Here’s a unique, I don’t know, unique to our use case situation that I came across with ChatGPT specifically recently. We were doing a live Basecamp, let me show you how this works, but there was data that we didn’t necessarily want to show. Chase was actually doing this. And so he did mock-ups of actual documents. He’s like, “Make me a PDF with this fake customer data, fake email addresses, fake phone numbers” so that we could show this is how it actually looks, but it’s not our account where you’re seeing these private things. And something as simple as that is like, okay, that just saved me from making it up and it did it in seconds.
Jason (07:21):
It’s huge. In fact, I did the same thing recently. I should have mentioned this example, too, I’m glad you brought that up. So I’m running Basecamp locally as we’re developing it. So I have my own local database and I can screw around with it without messing up the real thing. And so one of the things I’ll often do is I was working on UI on Campfire, our chat tool in Basecamp. I was working on the UI. I was curious, there’s a lot of stuff in there. I’d like to just get rid of it and move some stuff around or whatever. But the sample data, we typically have our seed data, which is like our default account that we use when we run things locally.
Kimberly (07:52):
Sure, like our demo type account.
Jason (07:55):
Yeah. It didn’t have a lot of chat history in it. And so I wanted to see a lot of different scenarios with file attachments and different people having conversations longer and shorter over many days and all sorts of different things going on. And we just didn’t have that in the standard seed data in our basic data. Now, historically, I would’ve asked maybe Merissa, and she’s done this, and I think you worked with her on this too, this incredible sample project stuff, which we have in the production version, but the local version, we didn’t have this. And there was a scenario, I just wanted to sort of paint this scenario, but to make it happen was very complicated because you have to log in as multiple people to chat with fake multiple people. You’ve been through it, you probably know. So I asked Claude, I’m like, okay, just in this one Campfire, in this one project, give me like three weeks of conversations between five different users, including file attachments and long conversations and short conversations and emojis and boosts and stuff, just as much stuff as you can that would feel like a real conversation over three weeks, about five or six different topics, whatever, something like that.
(08:59):
And I just said that and it populated the database and it’s like, it’s all done. I hit reload and it was pretty good, much better than nothing. Not as good as what you or Merissa would’ve put together, but I got it in three seconds or whatever. It was pretty instant. At that point though, I realized it was very repetitive and like too many one word responses. I’m like, this doesn’t work. Let’s undo that. And then do it more like every response should be at least 12 words and maybe a couple that are really short and whatever, thumbs ups and whatever. Anyway, after like a minute or two of screwing around, I just had a full Campfire chat with fake real people, with fake real conversations that helped me simply be able to design some ideas and see it in a bunch of different scenarios. And that alone is so hugely helpful. It’s like what Chase was doing, but with chat it’s especially hard because you have to come at it from multiple users.
Kimberly (09:55):
Right.
Jason (09:56):
Don’t want to just chat with yourself that doesn’t look like a real chat. So the fact that I was able to spin up fake data like that, fake real data was so incredibly helpful. And then you could also just be like, okay, I’m done, undo it all. And it just gets rid of it all. So it’s a sort of temporary, fast worker to do some things just to get a sense, and then you can back out. You can also leave it there if you want, or just back out and do something else. Hugely helpful for me, especially as a designer wanting to see things in a certain way, it’s hard to design when you don’t have the baseline data to design for.
Kimberly (10:29):
Yeah. And when I was talking to David about this, it sounded like he, more recently, not had come around AI, but as things have evolved more, he’s been like, okay, now I’m on board in a way that I wasn’t several months ago. Are you in that same boat where you’re more onboard more recently or have you always been pro AI in your everyday use?
Jason (10:53):
I would say, and I won’t speak for David on this, I think where he was coming from was a bit of skepticism initially with code quality and the whole thing.
Kimberly (11:02):
Yeah, agreed.
Jason (11:03):
And so then I think he was impressed by the fact that it had evolved because back three, four months ago, it wasn’t as good as it is today.
Kimberly (11:09):
Right. Even just like the end of last year.
Jason (11:12):
Yeah. So I don’t think I was skeptical because I hadn’t done a lot of these things that I’m doing now. So I think I’ve hopped on board, let’s just say. I mean, I was using AI personally for all sorts of like just replacing Google stuff, right? Mostly. And then at work, I was using it for some writing stuff, but now getting into product development, I’m just impressed by how quickly I can do certain things that I would’ve had to ask someone for before or really get up to speed to do again. So I wasn’t as skeptical and now I’m sold. I’m just like sold now.
(11:45):
I think Claude could have done this six months ago too. I just wasn’t really digging into it in that way, but I think I’m just riding the wave right now. A lot of people, of course, are talking about it and showing off what it can do. I’m like, well, I should kind of get in there and figure this stuff out. And it’s been very, very helpful. I’m sure there’s a million ways, if someone was sitting next to me here who really knew all the things you could do and all the ways you could do it, I’m sure I could learn a lot more. But for now, the things I’m able to do, I’m really appreciative that I can do those things. I’m using it in the ways that I find it to be valuable for me and I’m not searching for use cases. I’ve got the things it’s doing for me right now.
Kimberly (12:19):
Okay. So now let’s talk a little bit about our products. Of course, we’re not going to reveal any spoilers, but tell me a little bit about…
Jason (12:26):
We’re not, that’d be more fun, wouldn’t it?
Kimberly (12:28):
I mean, it would, but I figured you didn’t want to.
Jason (12:30):
I don’t really have any spoilers at the moment, but-
Kimberly (12:33):
Tell me what you’re thinking in terms of AI and our products. Is that something that as we’re launching Basecamp 5, you’re digging into? Is it a not now? And then also our other products, not just Basecamp, but HEY and Fizzy, kind of where are you thinking the next step for AI is for us?
Jason (12:48):
Yeah. I mean, we explored it quite a bit in Fizzy and I think there’s of course an endless amount of things you can do and maybe we should do at some point. Things we explored it for in Fizzy weren’t entirely useful at the moment. That said, I can imagine, for example, duplicate detection, like this bug looks like these three bugs. Is this the same thing? Things like that I think would be handy.
(13:10):
We had some summary stuff like summarizing a week of worth of work and what’s been going on and we just found those to be adequate but not interesting to read. And I think that that ultimately kind of pushed us back away from it for a minute. We can generate summaries, we can generate reviews of work, but like if no one really wants to read them, ahhh it didn’t feel right. So we kind of backed away from that. Anyway, that was a few months ago. David and I and Brian actually just caught up yesterday about AI and Basecamp 5 and we’re exploring some different avenues for inclusion of AI in the product. It’s also very, very interesting time because, and this is sort of David’s argument, which I buy, I also have other arguments that I’m trying to push forward, that a lot of people have spent a lot of energy building a lot of custom AI stuff into their existing products.
(14:01):
The alternatives in the market, competitors, alternatives, whatever, other people in our sphere basically all have AI features in the products at some level and they talk about them very proudly on their sites. It’s very obvious that it’s like pervasive now and we don’t talk about it and we don’t have anything and that’s been an intentional decision so far. And David’s point of view, and again, I agree with it and I also disagree with it in other ways, but I really agree fundamentally with it, is that things have actually changed so much in the past month, like with OpenClaw, for example, and 24/7 running agents, things are just perpetually running for you and the ability for agents just to log in as normal people, that people are going to end up bringing their own agents to our products and just have them be normal users, like just have your agent sign up for an account and then you can grant it access and you can bring all the knowledge it has.
(14:53):
It can learn everything about your product in no time at all. We’ve seen this already. We already have agents in Basecamp in our Basecamp account. We’ve invited our own. And so you can see that there’s a lot of custom work you can do to try to do those things or maybe you can wait a little bit longer and the game’s going to change where OpenAI, Anthropic, Grok, Gemini, all these things will be offering these always on agents. Just like OpenClaw is, OpenClaw is like extremely technical right now. You got to set up your own server or use a virtual server and it’s very complicated, but it’s a view into what things are going to be very, very soon. So what we can do is make the product simpler, clearer. We are working on some other stuff, some CLIs, command line interface for Basecamp that would make it easier for agents so they don’t have to use a browser.
(15:41):
We’re doing a bunch of that stuff too, but currently we’re thinking that people are going to be bringing a lot of their own stuff into Basecamp. However, there’s also stuff in Basecamp that we can be doing with AI that we’re going to be looking into doing for Basecamp 5 when we’ve already begun to think about these things. So I can’t reveal anything more than that, but I do think there’s going to be this hybrid world of some native AI features within a product and then people are going to bring their own agents that are connected to all sorts of other things also and that are going to know them really, really, really well and they’re going to bring those into the product as well, just as if they were coworkers. You’ve probably seen Jeremy’s doing this and David’s doing this and a few other people are doing this.
(16:17):
And it’s incredibly impressive. And it’s interesting because we didn’t have to build anything to make this happen. And that’s I think ultimately where the puck is going. That said, again, there are, I think, specialized things that we should be thinking about and things that are just more straightforward than having to think about signing for something somewhere else and then bringing it in versus just having a few things available to people. So it’s very exciting. It’s very much like on the edge of following what’s going on here and trying to determine what the best path forward is. And I’m actually very glad that we have not spent the past year building things that might be undone in a matter of weeks by a better way to do things. So there’s a point though, you can’t wait for someone else to invent the future that you want…
Kimberly (16:58):
Yeah
Jason (16:59):
because you might have to wait 18 months and it might be too long. So it’s figuring out the right timing for these things is always challenging. It’s very hard to call the top or the bottom of any market or any situation. At some point though, you got to just go, “Oh, this is very interesting. This is going to change. What else can we do though in the meantime that still is very helpful?”
Kimberly (17:14):
Yeah. I mean, it feels like you should always be early on the early side of things, but this is one situation where if you were early, you’re redoing basically what you’ve already done and with the small teams, we don’t have time to do that.
Jason (17:26):
Yeah. Look, we also have a small team. We have 62 people in the company. A third of the company might be engineers and designers really, ultimately. And AI is very helpful in allowing us to develop faster and fewer people in the whole thing, but ultimately we don’t have a company of a thousand people of which a hundred can go explore some of this stuff. So we still have to make decisions. There’s a lot of trade offs to make about what’s worth focusing on. And we currently think that the core features of our product, Basecamp, for example, that are ours that work the way we want them to work is a better place to invest most of our energy. We’ll still be investing some energy into AI and also have open arms and open doors for people to bring their own AI into Basecamp and make it easier for those AIs or for those agents to be able to access data in Basecamp. So we’re making that a lot easier. So I think we’re going absolutely the right direction here. Again, I’m glad we didn’t spend 10,000 engineer hours over the past year and a half or whatever it would be, building stuff that’s kind of going to be obsolete pretty soon.
Kimberly (18:24):
Yeah. It’s interesting because I see some of the customer feedback. People write in or comment on our YouTube videos that people are very polarized by this topic. I’ve said it before, but there’s people who are like, “When is Basecamp getting AI? You guys are late.” And there’s people who are like, “Don’t touch my Basecamp. It’s like the only piece of software I have that hasn’t been sucked in with shitty AI.” So it is one of those hard to balance between those different opinions and doing what’s right for the product.
Jason (18:53):
For sure. And we want to make both those sides happy, and I think we can. I absolutely think we can. Our approach is always to be as straightforward as possible, as no nonsense is possible. So slathering AI on everything everywhere all the time is not going to be our approach. And I’ve used tools that are like that now, where everything is like AI first. And I just think it’s a bit of a novelty at the moment. And I think it’s wearing thin in some ways and you just have to be careful not to sort of ruin things because not everybody wants that all the time. It should be available for sure. And others are very gung-ho about it, want it everywhere. So the good news is they’ll be able to bring theirs and have it wherever they want and do whatever they want. Meanwhile, for those who don’t want to go down that road or aren’t sophisticated enough or aren’t interested in enough in doing that, we have to provide some assistance for them as well and give them some leverage that they didn’t have before and show them that this stuff is very powerful and very useful, but also not in your way.
(19:48):
So it’s a delicate balance, but this is the same delicate balance we’ve been balancing on since the beginning of Basecamp. Everybody wants more stuff. This is the nature of software. Everyone wants more stuff and everyone has their two or three requests and they can’t understand why you haven’t done them yet. And then you’re like, well, there’s 85,000 people asking for two or three things. And some of those things overlap and many of them don’t. And then you also, people will also say like, “I love Basecamp because it’s so straightforward and simple and thank God because everything else is a mess.” And you’re like, big reason for that is because we’ve held back doing certain things that everyone’s been asking us to do and it’s always a delicate balance. So it’s no different. Expectations are no different. They’re just about different things. So it’s always the thing we’ve been really good at, I think, which is understanding the limits, making things very accessible for a huge swath of people and not getting ahead of ourselves and making things complicated to benefit a handful of people who really want the most and the many, and instead just kind of figure out what’s the right collection of things that makes the most sense for the most people that’s easy and approachable and understandable for nearly everybody.
(20:48):
And also there’s some more power around the corner if you really know how to get at it, you can do that. A good example of this, well, let’s call it Basecamp 3, actually all the way back to 2, kind of had a handful of tools that were, you could have to- dos and messages and files and documents and schedule stuff basically in a project. And that was the same up until a couple years ago, we added this feature to allow you to add multiple tools to a project so you could have multiple to-do sections. You could always have multiple to- do lists, but I give to do sections or multiple message boards.
Kimberly (21:20):
Two separate chat tools.
Jason (21:22):
Right. And so it’s still simple for everybody. There’s a collection of simple, straightforward tools that make sense to everybody, but those who want to reach around the corner and pick out a few more things off the shelf and put it on their project, they can, but it never gets in somebody’s way if they don’t want to think about the fact that they can do that. That’s always the line we’re trying to tow here. And so I think the same thing will be true with AI.
Kimberly (21:44):
Okay. A little off subject, but I’ve read a lot or heard recently about the whole tech surge of people being laid off because AI’s going to take over all these jobs and we don’t need programmers and we don’t need all of these different positions. I kind of wanted to get your take on that. One of the things I think of in particular that drives me crazy with companies, their use of AI, is support and offloading that support to AI robots where you can’t talk to a real person. I want to kind of get your take on the company’s philosophy when it comes to the roles that we have here and AI, how those might be supplemented or not.
Jason (22:23):
My longstanding opinion about most companies is they tend to have too many people. I think companies tend to be too big, teams are too large, and we’ve intentionally kept our company as small as we possibly can since we existed. We’ve gotten a little bit bigger than we are today, but we’re about 60 people, that feels like really good. And a lot of the people in our industry have teams of hundreds or thousands of people in their company. And I just never really understood that. Fair enough, whatever. So in general, I think that companies will often look for reasons to let people go that may or may not be true. At some point, when their numbers aren’t right and Wall Street, if they’re public, Wall Street’s demanding this or demanding that, it’s easy to lay people off. And you can say, “We’re laying people off because AI is going to make us more efficient or more productive.” And that might be true, but it also just might be an excuse to lay people off. So I don’t know. I don’t know.
Kimberly (23:17):
Sure.
Jason (23:17):
There’s no question that it should make people more productive, but I think people should have probably been a lot more productive to begin with. I don’t think you need a team of 12 people working on something that’s relatively small. So it’s all one and the same for me there. I agree with you that in some cases, AI-based customer service is actually quite good if you have a very simple, straightforward question. It’s also incredibly frustrating when you have something a little bit more nuanced and you just have this boiling urge like, “I just want to talk to somebody who will understand where I’m coming from, who’s not as intelligent as AI technically, but totally gets what I’m talking about because they’re human and they get it.” Right? Now, there’s also terrible human customer service out there as well.
Kimberly (24:01):
True.
Jason (24:02):
And that’s because people aren’t trained well and companies typically see it as a cost center, so they try to find the lowest common denominator. So it’s not that humans are better or worse at this. It’s like you want highly trained, long-term humans who know a product inside and out, are extraordinarily good, and that’s what we offer. We do offer some AI help in some places. You hit the little question mark and you can ask a question and stuff, but you just email support@basecamp.com and you’re getting a human.
Kimberly (24:32):
Right.
Jason (24:33):
So if we’re around in 245 years and we’re all dead and that might be different. In the short term, near term here, I think humans are hugely important. I think we have a massive advantage because we have incredibly good humans on our support team, many of which have been here for many, many, many years, some of which have been here for 15 years on support. This is a career job here, not just a temporary job, which it is the most place that’s kind of a temp job. And we’ve developed incredibly good people who have a huge depth of knowledge about how our products work and they really, really care. And I think it’s a massive competitive advantage for us and I would not want to give that up. That said, there’s also times you just want to get a quick answer. So we should have both and make both available, but it shouldn’t feel like you must go through the AI and then get pissed off enough to get a human.
(25:18):
I don’t ever want to have us do that. Those are the experiences I never want to have with other people’s companies. And by the way, it’s so different than old school support, like phone trees, you’re like, oh my God…
Kimberly (25:28):
Press five.
Jason (25:29):
And then the recording is so slow and you just…
Kimberly (25:32):
Agent. Agent.
Jason (25:32):
Slam the zero button or whatever and try to break through it. I don’t want anyone to ever feel like that with us.
Kimberly (25:40):
Yeah. The reason I brought it up is I recently had this terrible experience with an airline trying to get a question answered and talking to their AI bot, trying to get a real person like, okay, you’re not answering my questions, so I need to get to a real person. And literally being in a circle of, “No, I’m the bot and I can help you.” Well, you’re not helping me. “No, I’m the bot. I can help you.” I’m like, oh my gosh! Infuriating
Jason (26:01):
I know. Well, there’s this sort of know-it-all complex. And you’d imagine AI technically does know more than any human at this point, essentially, but it’s still different because people relate, it’s not just about knowledge. It’s about relating to somebody and inventing, frankly, to a human who understands that you’re pissed and can feel that and understand how to respond in a way that’s still just human to human. Humans want to talk to humans, especially when they’re pissed. They just do. I mean, you could also yell at an AI and let it all out and not worry about insulting anybody either, but I don’t think that’s what people actually want. I think when people really get a little bit frustrated and sometimes when they’re writing support, they are.
Kimberly (26:45):
Yeah.
Jason (26:45):
They are. And it might be because they had a bad day. It’s not that the product is terrible. They may have had a bad day or who knows what happened, right? They have a deadline coming up and they can’t find this thing they knew was there. And people are frustrated for all sorts of reasons. You want someone on the other side who can meet that, absorb it, understand it, and know how to work with you on it. Now, it’s not that AI can’t be trained to do that, and I’m sure there’s some very sophisticated models that can do that, but that’s a technologist’s point of view. If I’m a human, there’s a point, and it’s not a very deep point where I want to talk to somebody. I just still do. Maybe in 20 years I don’t. Today I do. I believe that’s true. I talk to our customers, I know it’s true, and we want to make sure that we’re never skimping on that.
Kimberly (27:31):
Yeah. Well, that seems like a perfect place to wrap it up. We will look forward to seeing what comes in the way of AI as we launch new products. This is a production of REWORK. You can find show notes and transcripts on our website at 37signals.com/podcast, full video episodes on YouTube. And if you have a question for Jason or David about a better way to work and run your business, leave us a voice recording. You can do that at 37signals.com/podcastquestion or send us an email to rework@37signals.com.