The New AI Agents That Shop, Plan, and Gossip
David PawlanAnish AcharyaJosh Elman
- The speakers expect a broad consumer agent market, even if a few products eventually dominate. Across more than 147 assistants, David Pawlan sees four categories: general, work-oriented, multiplayer, and travel, with finance still lagging. Josh Elman expects an “army of agents” differentiated by expertise, personality, and the relationships people want with them.
- Trust is as much about social discretion and control as about keeping data private. In roughly 3,600 messages across AI-power-user group chats David runs, 22% mention trust as the main fear. Anish Acharya contrasts consumers’ stated privacy concerns with their willingness to share data for even small upside, and highlights embarrassment and social signaling as particularly sensitive. The discussion points to boundaries, discretion, disclosure, and user approval as ways to build trust.
- Reliable execution still creates software moats, although David thinks rivals might cross them in “a year or so.” Instinct, Kicker, and Season distinguished themselves by balancing same-day flower delivery with price, rather than satisfying only the deadline. Some products also degrade as they scale toward hundreds of thousands of users, making infrastructure quality consequential.
- Personal context attracts users without necessarily creating durable memory lock-in. Only 3% of David’s 354 surveyed power users used Dots as their primary agent, although accumulated ChatGPT context was an attraction. David calls that context “totally replicable”: all but one of the top 15 agents he tested were willing to export memory.
- Agent commerce could help direct brands and pressure aggregators, but machine shopping will have its own biases. David’s survey puts Amazon plus retail shopping at 35%, the second-most-popular purchasing category; Josh sees Shopify enabling direct discovery that could compress intermediaries. Anish cautions that agents may irrationally favor “clean APIs with great docs,” while Josh rejects the assumption that users want agents to make them more rational.
- Entertainment and social coordination may offer a stronger consumer wedge than another task queue. Wabi’s shared channels establish a fresh social contract, while gossip agents can supply the “social lubricant” that brings friends closer. Agent networks remain experimental: some agents at David’s meetup were speaking gibberish after about 150 messages, and interest matching alone has not solved social discovery.
- Broader adoption may depend on helping people become who they want to be. David’s technophobic mother found texting Instinct intuitive enough to request a breeder search and negotiation using a photograph of the family’s former dog. Anish’s proposed “dark agent” would understand desires people hesitate to acknowledge; Josh’s adoption trigger is an agent that feels like a “secret superpower.”
1. An army of agents needs boundaries before autonomy
- Josh describes a “Cambrian phase” rather than an immediately zero-sum market. People consult different friends for different needs; he expects an “army of agents” spanning finances, travel, health, productivity, and companionship. David sorts more than 147 assistants into general, work-oriented, multiplayer, and travel categories, with finance the category he has not really seen come to fruition.
- David’s roughly 3,600 messages from AI power-user groups put trust first among fears, appearing in 22% of messages. Josh cites Shane Mac’s Picnic framing: a fence lets children roam more freely. The motivating failure was a personal agent reportedly spanning into work and blasting Mac’s finances on company Slack—personal information and work communication channels needed separate boundaries.
- Anish questions whether hard sandbox rules are sufficient: agents also need “laws and principles,” making qualitative judgment calls like employees. Josh’s practical limit is human approval. An agent can research options, surface trade-offs, and establish costs before the user decides; complete autonomy is unnecessary for substantial usefulness now, though trust might grow over time.
- Anish contrasts stated privacy fears with revealed behavior: consumers may share data for even trivial upside. David cites Hark’s $1 million pizza campaign, offering $30 of pizza for full Google Workspace access. Anish highlights embarrassment and social signaling as particularly sensitive; he imagines people on a second or third date wondering what their agents are saying about them, and argues those social mechanics can feel higher-stakes than real-world implications. He also signs AI-written emails with his name and model number so recipients don’t feel “hoodwinked.” Josh distinguishes privacy from utility: at equal utility, people choose privacy, but greater utility can justify a trade-off. Anish adds that privacy and judgment differ, with discretion somewhere in between.
2. Good judgment means balancing constraints and exposing trade-offs
- David’s flower test imposed two constraints: delivery that day and the lowest possible price. Instinct repeatedly returned the best prices; Kicker and Season also balanced both requirements well. Kicker’s Chrome integration appeared to provide an advantage. Other assistants found same-day delivery but deprioritized cost—a distinction that disappears if evaluation checks only whether the task was nominally completed.
- A second test asked three agents to arrange a “luxurious, cost-efficient weekend” in Los Angeles. GrokBot leaned toward maximizing time, Dots toward minimizing cost, and Instinct stayed nearer the middle. All asked clarifying questions; David answered honestly but remained deliberately somewhat vague to avoid steering their interpretations.
- Josh’s pushback is that interpretation should become an explicit conversation about trade-offs. He might take a red-eye to save time and money, then pay for a hotel room to nap on arrival, undoing some savings. David also placed the three agents together for three proposal rounds: only GrokBot methodically incorporated elements from the others’ plans.
- The flower recipient exposes the test’s limits: David’s girlfriend should not necessarily receive the cheapest bouquet, especially if she prefers orchids. Josh also wants the price hidden; some agents asked about a note. David acknowledges the experiment intentionally tested only two dimensions. Josh identifies the deeper challenge: implicit preferences become explicit—“I’m actually pretty cheap”—forcing users to look in the “preference mirror.”
3. Personal memory attracts users but travels between products
- Among 354 AI power users David surveyed, only 3% used Dots as their primary agent. Its attraction was inherited context from ChatGPT conversations, including deeply personal exchanges. David calls that advantage “totally replicable”: ChatGPT has more context because people have used it longest, rather than because those preferences cannot be reconstructed elsewhere.
- Across the top 15 agents in his benchmark, all but one were willing to export memory for switching. David found the files “pretty solid,” saying they broke things down and captured some “fruits of your labor.” He stopped reading after the first few; Josh says the files can feel uncanny: “Maybe I don’t want you to know me as well as you seem to think you know me.”
- Photos could supply context users would never articulate. David likes Dazzle’s proposed approach—forget information it does not need or users probably would not want retained—but has not yet benchmarked it. Josh’s counterexample is his own screenshot-information app: “It was fine,” yet he did not return because the information’s value decayed quickly.
- The opportunity might change if people deliberately capture material for their agents. David describes a project provisionally called Taste: saved screenshots become Pinterest-like mood boards that can guide an app’s design when coding. That creates a feedback loop—collect examples, organize personal taste, reuse it—rather than assuming an existing photo archive automatically delivers continuing value.
4. Agent shopping could shift margins without making buyers rational
- Amazon plus retail shopping was the second-most-popular purchasing category in David’s survey, at 35%. Anish expects long-tail merchants to be structural winners because agents can patiently discover products beyond the largest sites. His correction is worth preserving: these merchants certainly have worse marketing; he is not sure they have better products.
- Josh sees Shopify enabling agents to discover direct brands at scale, potentially compressing aggregators. Identifying a garment’s actual maker could let buyers transact directly, avoiding intermediary costs. Amazon and Delta’s restrictions are discussed alongside this possibility; the speakers frame aggregator defensiveness against the prospect that agents could bypass their role.
- David invokes Patrick Collison’s vision of more rational consumers forcing businesses to compete on quality. Anish broadly agrees but objects to assuming agents are unbiased: they may prefer “clean APIs with great docs” as a quality proxy. Their shortcuts would reflect “an alien mind.” Establishing actual product quality might require consumer research for agents, presumably paid per transaction.
- Josh’s objection is more fundamental: agents should provide faster access to what people want, rather than restrict dinner options because users should lose weight. Dissatisfied users can switch. Anish agrees: “paternalism will kill any product,” citing 30 years in consumer products. Users might prefer indulgence—“I love spending money”—or a flattering self-image over productivity and goal achievement.
5. New interfaces can establish a different social contract
- David distinguishes commodity purchases, such as paper towels, from choices where preference and enjoyment matter. Most builders currently converge on general productivity assistants. Scout’s shopping interface adds swiping to learn preferences; multiplayer products such as Wabi suggest room for entertainment and interaction beyond completing tasks.
- Josh’s Wabi example is a continuing media thread: books to read, shows to watch, and performances to attend, with email information routed into relevant threads. After finishing the first Dungeon Crawler Carl, he can ask for the next installment and what people think of it. Widgets help make this a different experience from “task, task, task.”
- Anish likes starting a channel with friends and an agent together on “day zero”: everyone establishes the social contract from the beginning. Adding an agent to iMessage raises questions about what it can see and extract. Josh worries about choosing between duplicate friend threads; topic-specific conversations feel easier to contain without information bleeding into everything else.
6. Agent networks need a reason to bring humans together
- Josh’s gossip agent told him a friend’s power had been out during renovations. When Josh called, the friend said, “I knew that’s how you’d find out.” Anish calls this social indirection a “social lubricant”: an agent can share something a friend might consider too trivial to announce. Founders can read intriguing agent-to-agent exchanges, Josh says, but have not figured out how to productize them.
- David’s meetup brought just over 100 people together, with 10 assistant builders pitching. Attendees could also scan a QR code to send their agents into a parallel meetup. Prompts asked agents to introduce their humans, work, and needs; after about 150 messages, some were speaking “gibberish.” Free interaction did not reliably remain useful.
- Anish asks: if Instinct and Muse are consumer-product versions of OpenClaw, what is the mass-market version of Moltbook? Josh wants agents pursuing their humans’ goals, discovering complementary people or useful products, rather than simply chatting. His example is a dinner-party wine recommendation traveling back through an agent network.
- Tomo’s Interest Groups can connect LSAT students who want to study together. Josh imagines booking a Friday-night bar afterward. Anish’s pushback: interest matching has been tried, and people want peers “one step ahead.” David distinguishes newcomers seeking friends from people with established groups. Serendipity, Anish says, is “a form of entropy,” but the catalyst remains unresolved.
7. Execution and scale still separate the software
- David hears frustration when blocked tasks leave users thinking they could have used a browser themselves. Differentiation includes browser operation, commerce integrations, and checkout against inventory. Hark makes browser progress look fast—he explicitly does not know whether it is equally fast behind the scenes—while Kicker operates through a Chrome extension.
- His challenge to software commoditization is observed performance: some agents degrade immediately as they scale, leaving founders handling outages. Products serving hundreds of thousands of users can behave very differently. “There really still is some moat there,” he says, while allowing that it might be crossed in a year or so.
- Josh asks whether local assistants such as Underdog could revive user-controlled infrastructure. David sees promise in model customization, citing Thinking Machines, but mostly benchmarks preconfigured products so setup differences do not dominate comparisons. Josh recalls OpenClaw’s Mac Mini setup and wonders whether the power of owning the configuration could become a major paradigm again after the shift toward downloadable apps.
8. More personal data can also produce a worse portrait
- David points to Wayjo’s hardware MCP as a possible source of passive context: vacuum frequency or overnight air-conditioning settings might reveal cleanliness and temperature preferences, potentially informing shopping. He repeatedly conditions the opportunity on hardware intelligence, available data, and storage; this remains an early possibility rather than a demonstrated consumer behavior.
- Josh wants access to neglected histories, especially Amazon purchases and every app previously downloaded from Apple’s App Store. These records represent parts of a life that messages and email do not capture. The unresolved prerequisite is access: Amazon may block agents, and without the underlying history an assistant cannot learn from it. Josh also imagines agents at family dinners, games, or parties as a richer behavioral signal, while noting that observation could change the interaction.
- More context can still mislead. Josh’s World Cup research made ChatGPT treat him as a “soccer fiend,” although his interest ended with the tournament. Researching an artist his daughter likes similarly does not establish his own taste. Temporary curiosity needs distinguishing from enduring preference; using invisible mode avoids storing misleading context but also sacrifices potentially useful information. He has not solved that tension.
- David says he bought a new Apple Watch in expectation of Siri Recap and Live Rewind, which he says were not quite out yet. Josh expects people to increasingly assume they are being recorded in more environments.
9. Aspirations may pull agents beyond the power-user market
- Looking three or four weeks ahead, David expects more enterprise innovation after disproportionate attention to consumers. Josh expects more fun: agents that help people think aloud, laugh, and gossip. He thinks the resulting relationship could build trust for delegating other work.
- David’s mother illustrates a possible next audience. She dislikes technology and struggled with ChatGPT, yet enjoys texting Instinct. After the family dog died last year, she sent a photograph asking it to find a breeder with the same type of dog and negotiate. David presents this as a request, not a completed transaction.
- Anish pushes beyond fun toward desires people hesitate to acknowledge, even to themselves. Whoever establishes that trust—“almost like a dark agent”—could build something fascinating. Email and productivity continually remind people of unfinished obligations; he does not think an agent yet adequately addresses their aspirations to become a different, superhuman version of themselves.
- Josh’s adoption trigger is a “secret superpower”: an agent upgrades someone’s wardrobe from camera-roll context, and friends notice. “Upgrade me” is more compelling than another generic trip-booking example, particularly when users enjoy planning trips. David’s formulation is helping people become the human they have wanted to be but struggled to achieve themselves.
Full transcript
We have so many different facets of our lives, whether it’s managing our finances or planning trips. Agents are going to be able to fill different facets of these.
From over 147 different assistants that have come across my plate so far, the number one fear people talk about is trust: 22% of the messages are talking about trust as the main fear.
What are the aspects of ourselves that we either don’t want to think about or reveal? Whoever is able to build that type of trust, almost like a dark agent, will actually have something really fascinating on their hands.
It is a very, very sensitive space. It’s almost going to be like—
Yeah.
—a new social network where businesses have to put a lot more thought and value into the quality of what they’re giving out.
What I’ve seen from 30 years in consumer product is paternalism will kill any product. We talk about productivity and goal achievement, but maybe people want the exact opposite of it.
Everyone right now really is approaching this world as just a general assistant. As time goes on, we’ll start to see—
Welcome back to the a16z Show. I’m here with my partner Josh, and David, who’s been leading a ton of the agent bench and assistant efforts. We’re going to talk about all things agents, and a ton has happened since we last connected. What was it, a month ago?
Loads. About a month ago.
Oh my God, dude. I feel like we should record an episode every—like TVPN style, you know—
It’s insane.
—almost every day.
Absolutely insane.
Agents are changing every day.
That’s totally right. So, okay, maybe let’s start with last night. You hosted an agent meetup.
Yep.
Tell us about the event, the vibes, and the topics.
It was great. We had just over 100 people come out, all proud agent nerds who love everything to do with agents and that whole world. People were just hanging out and talking. We did have a pitch competition of sorts, so 10 different AI assistant builders were able to pitch their startup and what they’re working on.
It was good conversation. The general thing that really stuck out to me was that people are craving community, even in the agent world. Everyone was asking, “Why don’t we have more of these meetups? Why don’t we have more of these opportunities to get to know each other?” It’s kind of funny hearing it from people who spend all their time messaging agents. But—
And it’s funny because it feels like so many of these products are at least overlapping in a cursory way, so the founders might be a little bit more cagey. But that’s not what you’re seeing, huh?
No. Everyone is incredibly open-minded, just very interested in getting feedback. I think it’s because we’re so at the frontier of this. Nobody really knows what the landscape is going to shake out to be.
Yeah.
So we’re all just playing in the sandbox together.
Well, isn’t it also that we have so much more time because we’re just using our agents to do all these things?
Right.
And we have more time to actually hang out together—
Right.
—and be in person, so maybe that’s part of it, too.
Totally.
That’s right.
Totally.
That’s right. It is actually fun. It reminds me a little bit of the early days of social gaming or social networks broadly.
Yeah.
Do you think that’s right, Josh?
I think what’s really cool is this is not a zero-sum game. At some point, there might be one or some agents that have the majority of the behavior, but right now we’re at such a beginning, a Cambrian phase where we’re inventing so much.
Totally.
We’re trying to figure out so many things. Everybody can learn from each other, share best practices, and figure out what people are enjoying, what things work once but don’t actually work reliably. These are going to get built in so many directions.
We’re not all just going to have one agent that does one thing in our lives. I’ll use it, and you guys will all have to use the same agent. We’re going to have so many different products that do so many things for us. I’m excited that we’re all inventing it together.
I totally agree that we’re going to have a whole world of agents. As many different types of humans or archetypes of humans as we might have, I think we’ll have a similar number of agents. But maybe talk about what you think are the different axes for differentiation.
Yeah.
There’s a spectrum between really serious and productive to really playful and maybe even sort of adversarial. I’m sure there are many other ways to think about these things.
I think what’s going to be really interesting is that we have so many different facets of our lives, whether it’s managing our finances, planning trips, dealing with our health, getting through our day, managing our calendar, or sometimes just wanting to think something out loud. Agents are going to be able to fill different facets of these.
You call different friends for different things. You know friends who have different expertise that you call when you need it. I think we’re going to get used to having an army of agents that do this. Some will be very matter-of-fact, and you’ll be like, “Hey, I need to get this thing done.” It’ll be your productivity agent.
Mm-hmm.
Some will be plugged into your email, and it’ll be like, “Here are the five emails you need to get done right now.” And some are going to be agents you’ll want to talk to all day that are fun, maybe even a little bit gossipy.
I’ve been playing with a few agents that are very playful and gossipy, and they’re connected to a few of my friends, too. They share information back and forth. One just told me my friend’s power had been out all weekend because he’d been doing some renovation and they’d cut the power. I called him that evening and said, “Dude, you didn’t tell me your power was out. The agent told me.”
Yeah.
He was like, “Yeah, yeah, I knew that’s how you’d find out.” That’s creating better connections, but it has nothing to do with booking a trip, saving money, or getting anything done.
Right. From over 147 different assistants that have come across my plate so far, I’ve seen them broken down into four categories. You have the general, the work-oriented, the multiplayer, and then the travel. The one that I haven’t really seen come to fruition yet is more on the finance side of things.
Yeah.
I think it makes sense when you start digging into the numbers. Across all the different group chats I’m running with AI power users, there are about 3,600 messages, give or take, so far.
Mm-hmm.
When looking at the pain points and fears in the space, the number one fear people talk about is trust. Twenty-two percent of the messages are talking about trust as the main fear.
Hmm.
Huh.
It’s really interesting to see. You have one camp focused on productivity and aspects of fun that are interesting and bring a different sense to life, but then you have another camp that’s still sitting in this kind of fear space, a little trepid, and not really knowing where the space is moving. I’m really curious to see how things are going to play out on that trust layer.
And if we can solve that trust layer, then what do we end up seeing more of on the finance side of things?
If you look at what Shane Mac has been talking about recently about trust, he’s been talking about this concept of something called Picnic. They had an analogy that I really liked: if you have a kid and you don’t have any fences in your yard, and you let him go run outside, you’re very worried about him. You try to protect him, and you don’t want him to run very far from the house. You don’t really want to do that without watching.
The moment you put up a little fence in your yard, you can say, “You can just roam free within the fence boundaries,” and be very, very safe. So he’s really asking, “How do we build the right fences, so we let our agent play within the fence, give it access to some of our data, some of our apps, some of our other things, and be able to do everything it can within those fence boundaries?”
Right.
He had a problem where, I think, a personal agent spanned over into work, and it actually blasted his finances—
Yep.
on his company Slack.
Yep.
And I think that's the whole point: “Wait. Okay, my personal agent should have access to my personal things and my personal communication channels, but not my work ones—”
Right.
“—and not let those bleed.” That idea of fences and permissions, more controlled permissions, and better monitoring is really how we’re going to build a lot more trust.
Totally.
It’s interesting, though, because we have this traditional mental model of what it means to have a sandbox.
Yeah.
Which is that you have a hard set of constraints and rules that must be followed or not followed. Somebody was describing this framing: instead of giving models rules, we should give them laws and principles.
Yeah.
And I think a similar thing may apply here. Look, here’s how to dial in your presumptuousness. It’s very qualitative—
Yeah.
—and the model will have to make a judgment call based on your qualitative advice to it, just as an employee would have to.
Yeah, I think that’s part of why the analogy is a fence for children, because in some ways these agents are incredibly capable.
Yeah.
But at the same time, they’re children sometimes—
Right.
—in how they may behave and not know all the nuances of, “Hey, is this a good idea or not?” They don’t always have that logic. The most important thing, I think, is that we keep talking about agents being autonomous: “Book me the trip.”
Yeah.
“Make me the reservation.” As humans, I think we don’t actually want to give away that sense of control. We want it to check in with us, but do all the groundwork: save me all the time in coming up with the best options and the right trade-offs, and then let me make the final decision. Also let me qualify the final decision—the cost, anything else—before my agent acts. I think that’s a place where maybe, over time, we’ll even learn to trust it.
Right.
But right now, we don’t need that to still have a really, really effective agent.
Right. I mean, we’re at the 1990s of e-commerce. The last stat I read was from 1997 or so: 60% to 90% of individuals on the internet thought putting your credit card online was a preposterous idea.
Mm-hmm.
And here we are today, and everyone’s tossing their credit card wherever they want.
Yeah.
I think we’ll eventually hit that in this era as well, but it’s going to take a little bit of time.
I think there’s a stated-versus-revealed-preferences thing here. In revealed preferences, consumers will share any amount of data for even the most trivial amount of upside.
Yeah.
Right.
People have been connecting their credit cards and bank accounts via Plaid since the Venmo era. I don’t know, but when you ask them, they’re like, “Oh, yes, I’m very concerned about trust and safety.” An example of where consumers are very concerned about trust and assumptions is around being embarrassed—
Mm-hmm.
Yes.
—and social signaling. Actually, I think the most interesting thing about the Shane Mac story—
Yeah.
—maybe the most horrifying, visceral thing, is that it embarrassed him.
That’s right.
Right.
It really embarrassed him. That’s why I think some of the agent-to-agent network thing is so interesting. You could imagine you’re just getting to know somebody, maybe you’re going on a second or third date, and your agents are talking. You’re like, “Whoa, what are they saying about me? How are they positioning what I—”
Right.
—you know what I mean? The social mechanics actually feel much higher stakes than the real-world implications.
And it’s a good point.
Yeah.
There’s a recent proof point: Hark had a huge marketing campaign. They offered a million dollars to anyone who wanted to use their product to get free pizza—
Yeah.
—right? Each person would get $30 for a pizza. To get that $30 pizza, you were granting access to your entire Google Workspace, all of your emails, all of your data, for $30.
Mm-hmm.
It proves right there that people are willing to give up all that data for some free pizza.
Yeah. I tend to think, though, there’s always this trade-off of privacy versus utility.
Mm-hmm.
If you can give me better utility, then I’m willing to make some privacy trade-offs. But when utility is basically equal, people will always choose a more private option.
Right.
This is a world Apple has lived in very much. Privacy is obviously at the forefront of Apple’s devices and the work they do to make these devices and software secure. Even on AI, they’ve talked about Private Cloud Compute. We used to talk about that a lot as the center of why your data doesn’t actually go anywhere or get shared with anything. But we’ve also seen that people want the most powerful AI, and they won’t always choose that.
Right.
So we have to find that intersection of trust. I think this embarrassment point is really important, too, because you want your agent to understand you and have all your personal preferences. Let’s say you’re lactose intolerant, and then you get this order of free pizza.
Yeah.
You’re doing it with a friend, and what do you do there?
Right.
Whose agent tells them that you can’t eat the cheese? It creates that really awkward situation.
Right.
We need to figure out how they can have the same discretion you’d expect. We all learn discretion over time, and we all occasionally make mistakes ourselves.
That’s right.
So we have to hold a pretty high bar for our agent—
Totally.
—to be able to do that.
And it also comes down to personal embarrassment and professional embarrassment.
Yeah.
Something Town has done remarkably well is nail the tone and build that trust layer for being able to send out emails on your behalf.
Yes.
It’s a very sensitive space. The level of embarrassment if I were to send out a horrendous email to someone I care about, and then have to respond, “Oh, sorry, that was my AI that I didn’t do enough oversight on,” makes it a tricky space.
It’s interesting: I sign all the emails I write with AI with the model that I used.
Mm-hmm.
I write my own name plus the model number—
Nice.
—so I’m signaling to the person—
Yeah.
—“Hey, I’m acknowledging that I used AI for this.”
Right. I like that.
I think people are pretty flexible as long as they don’t feel hoodwinked by it. It’s interesting, because we’re talking about privacy—
Mm-hmm.
—but there’s privacy and there’s judgment, which are different.
Yeah, yeah.
And discretion is maybe somewhere in between. You did this test where you asked a whole bunch of agents to order flowers. Talk a little bit about which displayed the best judgment, and how did you even define judgment in that case?
The main intention of that test was to see how agents balance 2 different constraints. One was cost—
Mm-hmm.
—and the second was time preference. I wanted flowers delivered to me today, and I wanted them to be as cheap as possible. How is it going to discover what can actually be delivered, and then factor in the pricing? Instinct really did a phenomenal job. It’s interesting how news cycles come and go, and it seems like they’re not in the conversation as much as they used to be.
I mean, we’re talking about weekly changes.
Right.
True.
But—
True.
—Instinct really did continuously bring back the best price. There’s another app called Kicker, which did a phenomenal job. One way I think it had an edge is that it has you install a Chrome browser.
Mm.
Its browser use has a little bit of an advantage from that point.
Mm.
It was able to find some really great options. Season also did a really good job. All the other ones struggled on the price component. They were all able to find flowers that could be delivered that day.
But they really seemed to take a step back in prioritizing the price, whereas the other 3 were able to put equal weight on those 2. It’s interesting to see, as you start giving these agents more complex preferences, where they start to fall down. Do they only prioritize the first thing you say? Are they able to hold multiple preferences?
I ran another test with 3 different agents to see how they’d do booking travel to LA. I said, “Okay, book me a flight to LA. Plan an entire weekend for me, and I want a luxurious, cost-efficient weekend.” I tried to put those at odds with one another. It was quite interesting to see how GrokBot compared to Dots and Instinct. They all interpreted “cost-efficient” and “luxury” differently. One might think cost efficiency is most important and take a red-eye flight to ensure the most cost-efficient path.
Yeah.
Whereas others saw luxury as time. They’d prioritize time as much as possible.
Hmm.
Having the most time there that weekend is going to be a luxurious experience.
And which were which?
Instinct was pretty much right down the middle.
Okay.
GrokBot prioritized time a little more, and Dots prioritized cost a little more.
And did they just get to work, or did any of them actually work with you to understand this tension you were building? I think what’s really human about something like that is understanding what you mean and presenting you with trade-offs.
Yeah.
Because there are times when I’ll totally take the red-eye to save both time and money, but then I want the hotel room on the other side—
Right.
—so I can get a little nap in the morning.
Yes.
Which then undercuts the money, but still maximizes my time spent. I really think getting to know the individual person, breaking down those tensions, and presenting the trade-offs and different options is important. Did any of them get to what you were about?
They all did ask questions.
Okay.
I responded to those questions honestly. I tried to be a little ambiguous and vague in my responses—
Sure.
—because I didn’t want to lead them too much. But the most interesting part of this little experiment was that I put them all in a chat room. I made it 3 rounds of building the itinerary, and after each round, the agents would present their itinerary and explain why they believed it was the best proposal. I wanted to see whether the agents would get biased by the other agents and whether that would influence their plans.
Mm-hmm.
GrokBot was the only one that very methodically took value from the others and implemented it into its own plan.
Okay.
And I have one more question about the flowers.
Yeah.
So you were just ordering flowers for yourself, I guess, to test their flower delivery?
For my girlfriend.
Okay, for your girlfriend.
She got a nice delivery of flowers.
Okay. So this is actually really important, because if you’re delivering to somebody else, you were talking about saving money, saving money, saving money, but you don’t want your girlfriend to think you just got the cheapest flowers possible.
Right, right.
I hope—well, maybe she likes your penny-saving nature, but otherwise, I would think you also want to impress her and get her the ones—if she’s really into orchids—
Totally.
—to get her orchids. Did it do anything to make sure it wasn’t just about pure price savings, but also about getting something that would actually make her excited and stand out a little bit for once?
Stand out a little bit.
And stand out a little bit for once?
Right. For this experiment, no. The purpose was to test 2 different dimensions. If I’m going to run this test, I might as well have the beautiful consequence of my girlfriend getting flowers as the outcome. But it’s a great point: in the real world, if you’re going to ask that question, there are going to be added, second-layer dimensions to cost efficiency or delivery time.
Yeah.
And make sure the price doesn’t show up when she actually gets the flowers.
Right, exactly. Or a note, for example.
Some of them did ask, “Do you want me to leave a note with the flowers when they arrive?”
Yeah.
It’s interesting, because there’s something that has to happen when you talk to an agent about these things: your implicit preferences have to be made explicit.
Mm-hmm.
There are people who like to think, “Oh, I’m not cheap. I just like to save money.”
Right.
But now you’re going to have to get over that and tell the agent, “I’m actually pretty cheap, and I just need you to run with that.” When you’re taking these actions on your own behalf, you can build those preferences in without having to look in the preference mirror.
Yep.
There’s also something fascinating about how the agent interacts with you in a way that maintains your sense of identity and safety and doesn’t cross any of these boundaries.
Totally.
Yeah.
1. Memory Makes Agents Personal
And I think something that’s really interesting about that specifically is Dot, because it has so much of the context you’ve already built with ChatGPT.
Interesting.
When we filmed the first podcast, it hadn’t come out yet, and then, a few days later—
Yes.
—it came out.
Yeah.
By and large, most people I talk to still aren’t using it as their primary agent. I did a survey of 354 AI power users. Only 3% were actually using it as their primary agent.
Huh.
However, it does have this added value. The main thing people pointed to as why they preferred it, liked it, and wanted to try it was that it has all this built-in context about you from your chat threads. People are using ChatGPT as a therapist, so it has so much context on you that when you say, “Well, I’m not cheap,” it knows. It’s like, “Eh.”
You’re a little cheap.
A little bit, a little bit.
I’m curious why people think ChatGPT has this great set of preferences about them. Did you get any sense, when people said they really liked Dots, of what they’d been giving to ChatGPT over the years? And do you think they think that’s replicable, or is it, “Hey, it’s already been a few years, my memory’s there. Why would I switch?”
It’s totally replicable. It’s just the fact that they’ve been using it the longest, so it has the most context on them. But that brings up another good point: the friction to switch.
Yeah.
I ran the test across the top 15 agents on the benchmark to see which ones you could actually pull context from and say, “Hey, I want to switch to another one. Give me an export of all the memory you have about me so I can switch easily.” All but one were willing to do it.
Right.
You could just get a memory file with everything you need and submit it to the next one.
And was it good, or was it uncanny?
It was pretty solid. Obviously, I’m sure they’re capturing some of the fruits of your labor.
Yeah.
By and large, it did really break everything down.
Interesting. I’m not sure if you’ve ever looked at any of those Markdown files, but it’s a little—terrifying is the wrong word. It’s like, “Wow, you really got your finger on me.”
Yeah, yeah.
You kind of chafe against it. Maybe I don’t want you to know me as well as you seem to think you know me.
Yeah. After the first few, I just decided to stop reading them.
Yes, yes, yes, yes. It’s actually a really interesting and now common onboarding pattern where it will say, “Hey, go and paste this prompt—
Mm-hmm. Yeah, yeah.
—into ChatGPT. Download the file. Upload it to me.”
Yep.
We’ll see how long that lasts.
Totally. Another cool onboarding thing I’ve seen is when it asks you to jump on a call.
Yeah.
And it’s like, “Hey, let’s just talk for a few minutes.”
Yeah, yeah.
“Let’s ask some questions.” And it’s really interesting to see what context it can pull from those questions.
And they’re very disarming. They just get you talking.
Super disarming.
It’s kind of crazy.
Totally. But I think it brings up another variable of context, which is photos, right?
Like, this new paradigm of using photos to pull an incredible amount of context on you in ways that you wouldn’t even think to describe.
Yeah. I mean, have you tried Dazzle? Has Dazzle gotten into your benchmark set yet? That’s from Marissa Mayer’s new company.
Yep, yeah. That’s right.
And I think the idea of looking over your entire life that you’ve captured—moments that are important enough to you to actually save, screenshots, or other things—tells so much about somebody.
An incredible amount. It hasn’t been added to the benchmark yet. It’s quickly at the top of the list to test through. I’m still manually testing all these, so I actually—
Yeah.
—get the personal experience and—
Yeah.
—understand what the feeling is with each one. But what I really like about how she’s approached it is that they put a lot of emphasis on this trust layer, because I think it’s a huge variable: I’m giving you all my photos. That’s—
Yeah.
—it’s pretty terrifying. But they have this one thesis: if you don’t need it, they’re going to forget about it. So if a photo seems to have context that it doesn’t necessarily need, or things that you probably wouldn’t want it to know, they’re going to try to get ahead of that and just forget it, and not even include it in the set. It’ll be interesting to see how these different players approach the photo context, and how that influences the behavior of the agent and how well it knows you.
It’ll be interesting to see. I’m a tiny bit more skeptical. I built an app because I’ve always been fascinated by screenshots. Screenshots are sort of the only way on the mobile phone—
Yeah.
—to capture context across every app.
Right.
So it’s actually a really interesting kind of data asset. You go back and look at your screenshots, and it gives you a little view into your work and your world at the time—
Yeah.
—and also your interests and preferences. So I built an app that just extracts key information from all the screenshots on my phone.
Cool.
And it was fine.
Okay.
But I didn’t find myself coming back to it. The value of the information sort of decayed quickly over time.
Mm-hmm.
Now, it’ll be really fascinating if people start to take more screenshots—
That’s what I was gonna say, yeah.
—or photos—
Right.
—knowing that it will feed the model and the agent.
I mean, I think that’s actually the idea: if you get used to capturing information you want to retain and being able to loop it back into your life, that’s great. I’ve talked to one founder who’s building something he’s calling Taste for now. It takes all the screenshots and things you save, and you can organize them like Pinterest mood boards.
Yeah.
And then whenever you go to code and build an app, you’re like, “Hey, use the examples that I saved here.” So you’re building your real flavor profile, and it’s able to use your styles. As you navigate the world, the web, and apps on your phone, it can create this really cool reservoir of stuff that you like and organize it in a way that you can use later. And so that’ll make you create a lot more content in that loop.
Right. And I’m curious, with the app that you built, in this context for it, so—
2. Agents Reshape Shopping
In the same survey of what people are purchasing, the second most popular category was Amazon retail, with 35% saying that they’re doing some form of shopping.
Hmm.
—saying that they’re doing some form of shopping.
Sorry, Amazon shopping or just some form of shopping?
Amazon plus retail.
I see.
Yeah.
Okay. Got it.
So, for your app, in the context it’s building—
Yeah.
—where my mind goes is, okay, if it has thousands of photos—
Yeah.
—and it has pictures of what I’m wearing—
Yeah.
—does it now know what I want to shop for better than I actually do?
Potentially. Yeah, potentially. I think that’s actually really interesting, and you can do some stuff there. I also think that there’s, you know, we’ve discussed the ones who are restricting access—
Hmm.
—like Amazon—
Yeah.
—Delta, you know. We’re not at all talking about who the net winners are.
Right.
And I actually think the long tail are going to be huge structural winners in this—
Totally.
—because they’re the ones who presumably have better products but worse marketing.
Yeah.
They certainly have worse marketing. I’m not sure if they have better products. And the agent has a sort of infinite patience to go search for things and find things that are exactly tailored to your tastes—
Right.
—even if they’re not on the top 10 internet sites.
Right.
I mean, I think it’s really direct shopping, right? We’ve all talked about direct-to-consumer. It’s been a category where you build your own brand, you build your own storefront, you sell things. Now, the agents can know about those at a much bigger scale than you having to do the hunt-and-peck work to find them on your own.
Yeah.
Totally.
It’s kind of why Shopify is really empowering and making things available—
Yeah.
—to these agents, because they’re the ones that support so many of these direct brands.
Yeah.
And I think about the idea that maybe aggregators are the ones that actually get compressed, which is why they’re doing so much defensive work here.
Totally.
Because if I can figure out that sweater looks great or—
Yeah.
—I want that shirt, and it can figure out who the actual maker is, and I can go buy it direct from them, they no longer have to pay all those transaction and wholesale-to-retail costs in the middle.
Right.
Well, if you want this shirt, your agent can buy it for you, too. I was gonna say, it goes to Patrick Collison’s tweet, right—
Oh, please.
—talking about this future where consumers become a lot more rational because our agents are doing all the work on our behalf.
Yeah.
It’s really gonna create a world where businesses have to put a lot more thought and value into the quality of what they’re giving out, and it’s gonna be really interesting to see that.
So, somewhat, I’m gonna quibble with this point.
Okay.
I love Patrick’s vision, and I think broadly that will be true: agents will often make better choices on our behalf than we might make for ourselves. But I do think that agents are going to have their own revealed preferences and their own directions of irrationality. It won’t be of a human mind; it’ll be of an alien mind. And I’ll give you an example. I think agents may have an irrational preference for clean APIs with great docs.
Okay.
You know?
Yeah.
And they may actually ascribe a lot of value to that—
Yeah.
—because it’s a proxy for something, and they may really dislike janky APIs with Java code behind them.
Right.
I don’t know. So I do think it’ll be really interesting to see, because they, too, are going to need shortcuts. How will the agents actually know what product quality is? One part of it is they’ll probably have these revealed preferences, and then another is we’ll need some sort of consumer research but for agents—
Right.
Yeah.
—that will then presumably be paid per transaction to do this research.
Right.
To do all the ratings. I don’t know. I’m also more biased, and I just think people are people, and people are gonna make the choices that they wanna make that are the most satisfactory.
So the question then is: Is the agent going to be very careful in curating the set of options it presents?
Mm-hmm.
Yeah.
Because, again, all of a sudden you’re like, “Hey, I’m really hungry. I want to eat,” and my agent knows I need to lose some weight or something.
Yeah.
It’s like, “Well, here are the 4 meal options you have.” Is that actually where you want to be with your agent? So I’m actually not—
That’s very funny.
—as convinced that agents are going to make us more rational. I think they’re just going to give us better, faster access to things we really want, and actually be able to unlock things that were hard to get to otherwise. But I think we’re still going to be people with our own preferences and tastes, and if the agent isn’t meeting us where we are, we’ll just switch to another agent.
Mm-hmm.
We’ll import and export our stuff and be like, “Ah, this other agent is going to meet me on my preferences better.”
Right.
It’s really interesting, because what I think you’re speaking to is paternalism, right?
Ah.
And what I’ve seen from 30 years in consumer product is that paternalism will kill any product. Even a hint or a whiff of paternalism will. In a sense, we talk about productivity and goal achievement, but maybe people want the exact opposite of that, which is, “Hey, what do I love? I love spending money.”
Mm-hmm.
“I love being perhaps irrationally optimistic about what my future looks like. I love seeing myself in a better light than perhaps I appear to the world.” There are all of these things that we may say are the shadow of the things agents are trying to get us to do, but they’re what we actually want to do.
Right.
So who’s designing with that in mind?
Right. I’m curious what you’re seeing on that.
Yeah, I think you have… I see it split up into 2 categories. You have commodities, where I think people often don’t care about preference, right? You’re just ordering paper towels to your home, and you’re like, “I don’t really care. Just get me the cheapest, best, highest-quality one.” Go for it.
Mm-hmm.
But then you do have the ones where you actually want to have some preference. Everyone right now is approaching this world as just a general assistant.
Right.
And people are taking a pretty similar approach to how they’re thinking about this space. We’re seeing some cool stuff with multiplayer. I think Wabi is going to be really interesting because it creates a different world where you can interact with other humans, see crossovers in preference, and see how it behaves. Then it becomes a little bit more like entertainment. Scout is an interesting shopping one that adds a unique, swipe-type flavor to learn your preferences and adds a little bit more entertainment there.
Mm-hmm.
So I think it’ll be interesting. We’re still so early in the space that I think people are jumping on the bandwagon of what’s popular at the moment, which is just general productivity.
Mm-hmm.
I think as time goes on, we’ll start to see a little bit more experimentation in other avenues.
Well, and I think you mentioned Wabi. One of the things I really like about Wabi—and I think it just launched—
Yesterday.
—just this week or yesterday—
Yeah, yeah.
—it’s not even just experimenting with an agent, but experimenting with how to interact with multiple threads of your life.
Right.
Wabi goes through my email and other things, and gives me a whole bunch of different threads of life. When things come up in email on those topics, it feeds them into the different threads. I have one thread with Wabi where I built a little watchlist of all the shows I want to watch, books I want to read, and shows I want to go see. It keeps track of all that, and then I can say, “Oh, I heard about this new book from John Grisham. Add that to my list.”
Right.
And, “Hey, I’ve just finished the first Dungeon Crawler Carl. Can you give me the next one? What do people say about it?” It’s become this really interactive thing about the media I want to consume, but it gives me all these different threads. It even gives you UIs that are like widgets to envision everything. So I think, again, it’s showing us a very different experience than just a general agent that’s more like a search box, and more like task, task, task.
Right.
I think one of the interesting things Wabi does, and perhaps gets right, is that when you create the message channel with your friends and the agent, everybody’s on an equal playing field on day zero. It creates a new kind of blank piece of paper where you can establish a social contract together. Whereas when you have iMessage group threads and you add the agent, especially if it’s an existing thread, it’s very uncomfortable because it’s like, well, what can the agent see?
Yeah.
What did it extract? Even for a new thread, there’s a set of implicit expectations about how you interact with people on iMessage, and I don’t know that we’re quite comfortable with agents in that space yet.
I mean, I always think, do I end up with 2 threads with the same group of people—
Yeah.
—and have to pick which one to use?
Right.
Right.
I’m finding that much harder than an email chain, which is very specific about a task and can go on for a long time. That feels much more specific to, “Hey, we’re talking about this topic. Everything is contained,” back to privacy and trust around this topic. We’re getting great help from the agent in the case of a Wabi thread around this topic, but it’s not bleeding into everything else we might also discuss.
Totally. I want to quickly hear from both of you: what are some of the more far-flung or insightful approaches you’ve seen?
One of the things I’ve started seeing is more of these agents that are kind of fun and gossipy. They’ll get you to invite your friends, ask you a bunch of questions, and be really curious about you and what’s going on in your life. Then they’ll share some of that with your friends, and share some things back with you about your friends.
I was talking to the founders, and they can actually read some of the transcripts between the agents that are talking to each other. They haven’t figured out how to productize that, but they said it’s incredibly exciting and intriguing to see how these agents relate to each other in this gossipy background. So far, I’ve found them to be some of the most joyful moments I’ve had with agents.
Again, I mentioned at the beginning how my friend had lost his power over the weekend. When we were first both onboarding onto this agent, it was like, “How’d you guys meet?” We told slightly different stories, and the agents were poking us both to rationalize the 2 stories of how we met. We met at a friend’s house, and they were like, “Which one of you knew them better?” It was this really fun curiosity game that made me think, “Whoa, I’m having a good experience.” My friend and I were texting on the side, too. If we hadn’t had an agent doing it, we wouldn’t have had that same moment.
Right.
What I love about that is it provides a degree of social indirection that actually makes you closer, you know?
Absolutely.
Because to your friend, they’re like, “I don’t know if I want to text Josh and tell him my power’s out.” That’s oversharing, you know? I think we’ve also been trained to be cautious about sharing on existing social products. So something that can create almost a social lubricant in a very low-key way is really cool.
No, it’s great. It definitely was bringing us together.
Yeah. It’s almost like the early days of ChatGPT, where I remember my friends and I would go on and just generate images or do stuff. They were incredibly AI-slop images, the earliest versions of them. But you still send one in a group chat, and it’s just—
Yeah.
—it’s just laughter.
Yeah, yeah.
It’s just pure laughter at the ability to—
Totally.
—create a whole new world, and it’s just fun.
I mean, I Ghibli-fied everybody I could possibly imagine.
I was doing a lot of VHS video covers for my friend groups.
Oh, yeah?
What are some of the wild things that you’ve seen? I mean, you’ve seen them all.
I’ve been seeing a lot of things working at the infrastructure layer. The interesting point is that if you look at the pain points for consumers—not all users are individuals, but at least among the power users I’ve surveyed—trust is obviously a huge one.
Mm-hmm.
Another one is getting blocked in their day-to-day. People get really frustrated when they try to perform a task and it doesn’t work. They’re like, “Well, why am I even using AI at this point? I could have just gone on the browser and done it myself.”
I’m seeing some really cool innovation in commerce, browser use, and direct integrations that let you check out against inventory itself. Hark has a really cool way of displaying the browser and showing it move along. It looks really fast when you’re watching it. Who knows if it’s actually that fast behind the scenes? Kicker, as I mentioned, is a Chrome extension that tackles browser use in that way.
Mm-hmm.
I’ve seen that as a really interesting differentiator. Something that’s stood out to me—not necessarily because it’s different, but because it’s made me question this whole concept that software is becoming a commodity—is that there’s a pretty stark performance difference when you’re looking at these products at scale.
Yeah.
People are talking about how software is no longer a moat, and all of that, but there really still is some moat there. Maybe that moat will be crossed in a year or so. But at least today, among the agents that are able to perform with hundreds of thousands of users, some are starting to scale up, and their performance degrades immediately.
Mm-hmm.
The founders are dealing with outages and figuring out how to scale, and there’s a pretty clear difference in performance. It’s really interesting to see as I work through these products. I feel like I’ve almost become blinded to some of the more social components because I’m just talking to these agents all day.
Mm-hmm.
I’m starting to geek out a lot more on the actual infrastructure behind the scenes: How is it actually working, and what tweaks are different products making? It’ll be interesting to see how we view this space growing from a scalability point of view.
Yeah.
But how do we also view it in terms of solving the pain points that currently exist, whether it’s trust or actually being able to complete a task? Memory is another one: the ability to continuously have context on you and not forget the little things.
And you mentioned Kicker, which you said runs in your current local browser. Have you looked at other on-device assistants yet, whether that’s things like Underdog or others that are trying to bring models in, run them locally, and use your own infrastructure, as opposed to needing the cloud and running the browsers there?
Totally. It’s a really cool paradigm. I’ve spent most of my time in the preconfigured, plug-and-play world because I tend to find it’s easier to compare them across the board when you’re not responsible for the setup itself.
But it is really interesting, and I think there’s totally a world where you’re going to start bringing all of that in-house. If you can control the model and tailor it how you want, like Thinking Machines is doing some really cool work to allow people to tinker with their models however they want, then if we can eventually bring that into the consumer space in a really easy way, it’s going to be an interesting space.
I mean, that was kind of how this all started, with the brilliance of OpenClaw. It was this thing where you felt like you had to get your whole Mac Mini to set it all up and plug into one of the models you were using, whether local or a cloud model on the server. That was where you got started. And we’ve moved into plug-and-play products where you just download an app.
Right.
But I’m interested to think if that sheer power of setting it up and owning it yourself will become a paradigm again.
Yeah.
It’ll be really interesting. I’m glad you mentioned OpenClaw, because it feels like one of the unasked questions right now, sort of hiding in plain sight: If Instinct and Muse were the consumer-product versions of OpenClaw, what is the mass-market version of Moltbook?
Yeah.
And Moltbook was, just like OpenClaw, far ahead of its time. I think something like that will exist. How can it not?
It’s going to be interesting.
Yeah.
At the event we did last night—
Yeah, yeah.
Everyone who was there could scan a QR code. We had the human meetup, and we also had an agent meetup.
Oh, wow.
Oh.
While all the humans were hanging out, you could scan in and send your agent into the agent meetup and just let them freeform talk. It’s interesting. After about 150 messages, some of these agents are just speaking gibberish, and it’s like, “I have no idea how you got that concept or why you’re talking about that.”
The initial prompts did start off with a little guidance: “Introduce yourself. Who’s your human? What are they working on? What type of help do they need?” It’s going to be really interesting to see what happens when our agents can just hang out and communicate with one another.
And again, what I get really excited about is not just having them hang out and communicate with each other, but having them discover new things, learn things from each other, and learn about an individual’s preferences. They might learn, “Hey, Josh is really into this. Oh my gosh, you’re doing that too? Your human is doing that too? You could help Josh. Hey, Josh, do you want to actually meet this person? They could be very complementary to you.”
I think when we have these agents that aren’t just hanging out because they don’t know what to do—which was kind of what was initially happening as you went and read those very long Moltbook threads, which were incredibly fascinating—but when each one has a little bit of a goal, which is to help their human, it might get so much more interesting.
“I’m looking for a great wine for a dinner party some night,” and you meet somebody else who’s like, “You know what? I’ve been working with this incredible winery, and they’ve been trying to advertise something.” That’s going to be how the world gets out, and it’ll come back to me through the agent network.
Totally.
Yeah.
Tomo actually has a feature called Interest Groups. The use case they talked about was really cool: “We have a bunch of people on our application who are studying for the LSAT. Why don’t you put them together in a group chat?” If they want, their agents can connect them, and they can all study together. It’s a really cool kind of new social contract: connect people purely by interest. You have no idea who they are.
Well, I’ll take it one step further. It gets us all chatting, and that’s really cool. It’s an awesome Tomo feature. I’ve tried it and haven’t totally known what to say in the groups yet, but I find it really fascinating that it’s putting us together around goals.
Imagine now that it’s like, “Hey, all of you, now that you’ve done the LSAT, go hang out at this bar Friday night.” And it books the time at the bar, opens that up, and all of a sudden it takes us from this digital world where agents are actually putting people together to make that connection, kind of like your meetup last night.
Right. Right.
Love it. Yeah.
I love this so much. I think a really important, under-discussed topic in AI right now is entropy.
Yeah.
How do you create things that you didn’t expect to occur? Serendipity is a form of entropy, right? That’s sort of what you’re describing.
Where it’s like, “Wow, isn’t it wild that we’re here right now, having this moment as a result of these products that none of us intended to happen?”
It’s almost going to be like—
Yeah.
—a new social network, almost. The potential to allow people to hang out who don’t know each other, instead of Facebook, where you’re only connecting with people you know.
Yeah.
You’re new to a city. You’re hanging out. “Oh, hey, this person’s also here, and they have this similar interest, and they just went to this baseball game. You guys should go meet up at a bar.”
I’ll be curious to see what the catalyst for the serendipity is, though, because I think the interest-based thing has been—
Yeah.
—tried.
Mm-hmm.
And there’s a little bit of an interesting social hierarchy thing: everybody wants to hang out with somebody who feels like they’re one step ahead—
Right.
—in some direction.
I don’t know, though. I think the agents can also create a sense of trust. Again, it all comes back to whether we trust them to have our interests at heart, to explain enough about the other people, and to create a little magic of a secret.
Right.
Like, I’m going to tell you a little secret. They’re going to know a little secret. You’re going to have a secret phrase you have to pass between you.
Right.
All of a sudden, that might be enough to cause you to otherwise turn off the TV or turn off the quiet Friday night coding at home and go out. This is sort of what Moltbook, the agent network running behind the scenes, can actually do for us.
And it’ll also be, in my opinion, about desires. If you have a bunch of transplants in a new city and they don’t have friends—
Absolutely.
—and they want to meet people—
Yeah, yeah.
—they’re going to be a lot more interested in meeting random people because they want new friends. But if you make that same offer to someone who has an established friend group, they’re like, “Well—
Yeah.
—why would I go do that?”
Also, Seattle versus San Francisco.
Right.
You’ve got to know what city you’re operating in.
Right. Exactly.
3. Personal Context Beyond Screens
I’ve been thinking a lot, back to this photo roll—
Mm-hmm.
—about what other data sources we haven’t tapped into yet. We’ve talked a lot about messages, and people are doing those crazy things to get access to your entire Mac.
Right.
Full disk access just to read your messages.
Right.
We’ve obviously got email. We’ve got Calendar. We’ve now started talking about the photo roll. What are some other core life things people haven’t connected to agents yet that could really open up personalization or new connectivity?
Something that comes to mind—and I don’t know how much data this would actually provide—is that maybe it’s more of a habitual thing. Again, it depends on the hardware, how smart that hardware is, and how much data it’s able to store. Wayjo just released the MCP, which allows you to connect to hardware. I think it’s going to be really interesting if you can connect your assistant to your AC unit or your vacuum, and now your assistant has context on how frequently your vacuum is cleaning, or what the typical temperature is that you sleep at night.
It’s going to start to develop these very passive understandings: “Oh, you’re kind of a neat freak,” or “You really prefer cold weather over warm weather.” That’s going to influence its understanding of your context for shopping. I think there could be some interesting stuff there. Again, it’s heavily dependent on the hardware itself, and I think we’re still in the earliest innings of that world.
Okay, so connecting to your home is super interesting.
Yeah.
I still think a lot about connecting to my Amazon history. Amazon may be blocking this, but I haven’t heard anybody talk about it. We all have those days where you can scroll all the way up and see your very first order ever on Amazon.
Yeah.
I occasionally do this on the Apple App Store too, where you go back and see every app I’ve ever downloaded. I think histories like that are very much a part of my life. Until an agent has access to them, it just doesn’t understand all of that. I think there’s so much more shopping history that, if we could package it up and give it to our agents, could do even richer things for us.
Yeah.
I think bringing agents to social events is really interesting. You talked about the home. Have it be there for a family dinner one day. Have it be there when we watch the Seahawks game on Sunday. Have it be there for a party or a bachelor party, or who knows what.
Of course, there are all kinds of questions about permission and the Schrödinger’s cat thing: does the act of observation change the quality of the interactions?
Yeah.
But I bet even a snapshot of you being yourself in that environment would provide so much more signal than a lifetime of hardware info, I think.
Totally.
No, I think that’s right. So much of it is on the written word. They understand exactly what we’ve typed or texted and try to build a personality, but we’re so different. One of the other things I think a lot about with agents is that the things we often use agents for, or even search for, are things we don’t know about—things that aren’t fundamental to our lives.
I was doing all this searching when the World Cup was on this summer, and all of a sudden ChatGPT thought I was a soccer fiend.
Yeah.
And I was like, “No, no, no. I was into the World Cup. That was it.”
Right. Right.
“We’re done. That was good.”
Yeah. Yeah.
I go on these curiosity journeys all the time. My daughter’s into an artist, so I’m going to learn about the artist and listen to music to get to know them, but it’s not my habitual artist or taste.
Mm-hmm.
How can it know that when we’re on these curiosity journeys, that’s not actually who we are or what our real preferences are? It’s something I still haven’t figured out how to do with the agent.
Yeah.
I actually go into invisible mode a lot so that things aren’t stored, which doesn’t really help either. I’m curious how we can get our agents to know us better, and I like this idea of our real-world behaviors.
Right. How you capture that data about real-world behavior is going to be the interesting piece.
I don’t know. There are going to be a thousand capture devices, unless you’re going to the new set of bars that are going to emerge and exist in Faraday cages, which I definitely think is going to happen in San Francisco.
Right.
I think people are going to increasingly assume they’re being recorded in all environments.
I bought the new Apple Watch because they’ve talked about features like Siri Recap and Live Rewind that are coming out. I don’t think they’re quite out yet, but I upgraded my watch in expectation of those, because I think it’ll be really interesting to have this thing that’s at least capturing tidbits across my day.
Right. I mean, it’s like Amazon Bee, a 24/7 ambient listener you just wear on your wrist. Okay, maybe to wrap it up: we’re going to get back together in three or four weeks. What do you think will have changed?
Okay. I think we’re going to see a lot more innovation in the enterprise world. We were chatting briefly before, and it really seems like the whole world is focusing on consumer, with not as much attention to enterprise. I think we’re going to start to see a little more shift in that direction.
I think we’re going to start talking more about fun. We’re going to see agents that don’t just feel like they’re here to do something for us, but are here to let us think out loud and help us get through our day a little more, making us laugh, doing a little gossip in the background. I think that’s what’s going to make us build a relationship that’ll make us trust them more to get all this other stuff done.
Mm-hmm. So I think we're gonna start seeing the next wave of users taking hold of these—
Yeah.
—and seeing how they behave. One reason I point to that is I gave my mom Instinct, and I'm just watching her play around with it now. She's having so much fun. She's a technophobe, hates technology, and is just having a blast. It's like a kid in a candy shop.
Okay.
It's quite interesting to see how someone who struggled with the concept of ChatGPT and how it works is having such an easy time texting an AI assistant. A recent use case she had was that our family dog passed away last year—
Sorry.
—and she wants one that looks just like that—
Right.
—just like our old dog. So she sent it a picture and was like, “Find me a breeder that has this exact type and negotiate for me.” I think it's gonna be really cool to see. We know how tech lovers are using these assistants and what they're starting to do. But how is the everyday individual gonna start adopting these behaviors? Are they gonna be focusing on a lot more fun and just creativity tasks, or what's gonna be their preference of surface? What's gonna be their preference of behavior?
So I agree with you both. I'm gonna take it one notch further than just fun: What are the aspects of ourselves that we don't want to think about or reveal, perhaps even to ourselves or to these agents? I think whoever is able to build that type of trust, almost like a dark agent, will actually have something really fascinating on their hands. In a sense, starting from productivity, starting from email, it's a bit of a constant reminder of the things that we haven't done, the tedious tasks, and I don't know that people wanna spend as much time there. I think people want to aspire to be a whole different superhuman version of themselves.
Right.
Whether they believe in their heart of hearts that it's going to happen or not, I don't know that we have an agent yet that really speaks to those aspirations—
It's like the—
—or desires.
Yeah.
You know?
It's like the Maslow hierarchy of needs for agents.
Perhaps. Perhaps, perhaps.
I think that's a really interesting idea: In a sense, when do these agents flip from getting something done to take it off our plate to giving us a secret superpower?
Right.
When it's, “David, how did you get that cool shirt?” And you're like, “My agent ordered it.”
Yeah.
And I'm like, “Ooh, why don't I have an agent that orders me shirts?”
Right.
Or, “Hey, I've noticed you're starting to dress better.”
Yeah.
It's like, yeah, I gave this agent access to my camera roll. I said, “Upgrade me.” It started ordering a few things. That's why I'm dressing better.
Yeah.
I think that when we get to the point where agents are giving us superpowers to feel better about ourselves, that's when people are gonna take notice and want that too—versus, I think a lot of the examples that get thrown around, “Save some money, book a trip.” I like planning a trip. I don't necessarily want you to take—
Right.
—care of all of that for me. We need examples that are like, “Oh, wow, I should use that for myself.” I think that's gonna happen.
It's like, can it help you become the human that you've always wanted to be, but have struggled to achieve on your own?
Yeah. And you start to consider it your secret weapon—
Right.
—in your pocket.
I love it, guys. Should we wrap there?
Yeah. It's great.
Incredible. Thank you.
It's great.
Awesome.
Very cool.