Instinct Went From $2.5B to $10B in a Month on $0 of Marketing. Noah Shinn, 23, Explains How.

1. Key Themes

Scarcity-Driven Referral Growth Beats Paid Marketing

Instinct's growth engine is a capped invite system where every send costs the sender something and doubles as an endorsement. Compounding started slowly (1-2% daily) and accelerated once the base passed a couple thousand users and people began posting their own use cases.

"We rolled it out to 200 people. Next day it was 205, and then the next day it was 210."

"Every user gets 5 invites, so every send costs the sender something and reads as an endorsement. People email Shinn asking for one, others brag about the 3 they still hold, and invites have sold on eBay for around $300."

"Rationing access beats waking up with 10x the users and 80% of them locked out for lack of compute, he says."

Trust Is the Core Metric and the Moat for Agents

Shinn treats the first credit card or password handed to the agent as the key measure of trust. Trust accrues over weeks, sharing is user-paced and revocable, and it creates a flywheel: more access makes the agent more useful, which earns more access. The article notes retention is high among users who share sensitive items, but cause and selection aren't separated.

"3 weeks in, there's a 40% chance that the user has shared a personal credit card with Instinct."

"The loop feeds itself. The more a user shares, the more proactive and useful Instinct gets, and that gives them a reason to share the next piece."

"Users who connect at least 1 sensitive item show an 80% retention rate, though the interview never separates cause from selection."

Compute Procurement Is the Central Strategic Constraint

At 10% daily growth, compute demand doubles roughly weekly, turning capacity planning into a high-stakes bet. Lead times of months and 3-4x premiums for short-notice buying make the order size a core strategic decision (Shinn spends about 40% of his time on it, per the article's section header). A proactive, self-waking agent also consumes far more compute than a prompt-driven one, so cost control relies on deadline-flexible work and efficiency stacking.

"What does it mean when the amount of compute that you need access to is now doubling effectively every week?"

"Capacity takes several months to arrive, and buying on short notice costs 3 to 4x."

"Instinct has the ability to wake up and to sleep at any moment in time during the day."

"Work that can finish in minutes or hours instead of hundreds of milliseconds runs on deployment shapes he puts at 3x to 8x more efficient on the same compute, and gains of 30% here and 10% there stack on top."

Agent Safety Requires Independent Oversight and Objective-Based Design

For agents that hold payment methods, Shinn argues safety must sit outside the agent. Instinct uses input firewalls plus a decoupled monitor that can intervene before any action fires. The agent is also built around standing objectives rather than raw prompts, which makes it harder to weaponize with ill-intentioned requests and extends its time horizon to months-long goals.

"It is validated and scrutinized by something that is decoupled from the same incentive system as the underlying agent."

"The first version shipped with neither, and once the team saw the gap it rebuilt the architecture instead of patching it."

"With most other AI products, you write a prompt and then it does the task and then it tells you what happened."

"Users already hand it goals that run 3 to 4 months, like a mile-time target, and he describes small businesses running their back office on it with standing rules such as keeping inventory above a floor and below a ceiling."

Agent Monetization Shifts From Subscriptions to Merchant Take Rates

Instinct is free to users and plans to monetize through a platform-wide transaction take rate, positioned as an exchange of distribution for merchant access. Shinn benchmarks against Shopify, Amazon, and Apple, with travel as the wedge. The article flags that volume is not revenue.

"I see a blanket transaction take rate being enforced across the platform, which is just us exchanging distribution for being able to serve products on behalf of merchants."

"He lays the rates out as a curve: Shopify at roughly 2% to 3%, Amazon at upwards of 10%, and Apple's in-app cut at 30%."

"Shinn says Instinct is approaching $1B a year in transaction volume on a very small invite-only base, with 50% of it in travel. Volume is not revenue, and reports on the round cite no revenue figure."

2. Contrarian Perspectives

Understandability Over Capability

While AI launches compete on capability lists, Shinn argues consumers are fatigued and don't know how to access any of it. He credits a simple rule for the engagement and word of mouth, and applies it down to message structure.

"Let's not focus on capability. Let's only focus on understandability."

"Most people read about 80% of the first line and 50% of the next before attention tapers, so he wants the point inside the first 30%."

Skip the $100/Month Subscription and Go Free With Take Rate

Conventional wisdom would monetize an agent via subscription, which would pay sooner. Shinn treats it as a local optimum and uses venture capital to avoid it, aiming for a larger outcome through merchant economics.

"A $100-a-month subscription would pay sooner. He treats it as a local optimum, and says venture capital is how he skips it."

"Free for life is not a promise he'll make yet, but free for everyone is his personal goal."

Incumbents Should Test Agents on 1% of Users Rather Than Fear Them

Rather than treating agents as an existential threat or ignoring them, Shinn argues the risk is manageable via small-slice A/B tests. His distinction: service-weighted businesses may see volume rise, while attention-weighted businesses face the hard case.

"You can mitigate the risk. You can scale down the experiment. You can run A/B tests to figure out if we enable this certain thing across 1% of users or something like that."

"Cut the clicks between a user and a ride or a meal and, he argues, volume rises because the same good gets easier to buy."

3. Companies Identified

Instinct

• Description: Consumer personal AI agent with its own phone, computer, and email; proactive, objective-driven.
• Why mentioned: The subject of the interview; valuation went from $2.5B to $10B in about a month on $0 marketing.
• Quotes: "On September 28, Instinct raised a $1B Series C at a $10B valuation from Sequoia, Benchmark and Coatue." / "Instinct gets its own phone, computer and email address, so you can text it or call it, and it calls you back when a deadline gets close."

Sequoia, Benchmark, Coatue

• Description: Venture investors in Instinct's Series C.
• Why mentioned: Led the $1B round at a $10B valuation.
• Quotes: "Instinct raised a $1B Series C at a $10B valuation from Sequoia, Benchmark and Coatue."

Opal (Opal Zero)

• Description: Agent access-management product (sponsor).
• Why mentioned: Sponsored placement on scoped, expiring agent access; design partners include Databricks, Faire, Elastic, and Superhuman.
• Quotes: "Opal Zero takes the keys back the moment the job is done" / "Built with Databricks, Faire, Elastic and Superhuman as design partners."

Shopify, Amazon, Apple

• Description: Commerce and platform incumbents.
• Why mentioned: Benchmarks for the take-rate curve Instinct could climb.
• Quotes: "Shopify at roughly 2% to 3%, Amazon at upwards of 10%, and Apple's in-app cut at 30%."

Claude Code / Codex

• Description: Coding agent products.
• Why mentioned: Comparison point for compute growth pace and for the maker-checker pattern.
• Quotes: "That pace outpaces what Claude Code or Codex faced, he says." / "The monitor is the maker-checker split from coding agents, pointed at an agent that spends money."

OpenClaw

• Description: Project referenced in a linked related article.
• Why mentioned: Cited as a growth-loop reference.
• Quotes: "Peter Steinberger Almost Deleted OpenClaw. Then 4.7 Million People Downloaded It."

Lovable

• Description: AI app builder referenced as a growth case study.
• Why mentioned: Related read on near-zero paid ads growth.
• Quotes: "How Lovable hit $400M ARR in 14 months with 146 people and almost zero paid ads"

Magnific

• Description: Referenced as a bootstrapped growth story.
• Why mentioned: Related read on distribution with no ads or VC.
• Quotes: "Zero Ads. Zero VC. $230M ARR. The Story of Magnific"

Opus 5

• Description: Frontier model used as a performance comparison.
• Why mentioned: Shinn claims Instinct matches it on key measures at low cost.
• Quotes: "Instinct matches Opus 5 on engagement, A/B tests and internal evals at a very low cost, per Shinn. Hold that as his claim until someone benchmarks it."

4. People Identified

Noah Shinn

• Description: 23-year-old founder of Instinct.
• Why mentioned: Interviewee; details growth, compute, safety, and monetization strategy.
• Quotes: "Noah Shinn is 23, and his personal agent, Instinct, adds about 10% more users every day." / "'we spent $0 on marketing so far.'"

Patrick O'Shaughnessy

• Description: Host of Invest Like the Best.
• Why mentioned: Conducted the interview; reacted to the trust data.
• Quotes: "Shinn walked Patrick O'Shaughnessy through the growth, compute and take-rate numbers behind that valuation." / "The host called it crazy for consumer technology."

Ruben Dominguez

• Description: Author of the newsletter piece.
• Why mentioned: Watched the interview and distilled the takeaways.
• Quotes: "I watched the full interview so you don't have to."

Peter Steinberger

• Description: OpenClaw creator (via linked article).
• Why mentioned: Subject of a related growth-loop read.
• Quotes: "Peter Steinberger Almost Deleted OpenClaw. Then 4.7 Million People Downloaded It."

Sam Altman, Sarah Friar, Satya Nadella, Marc Andreessen

• Description: Tech and AI industry leaders referenced in linked related reads.
• Why mentioned: Cited in connection with compute bets, compute scarcity, agents needing their own computers, and AI moats.
• Quotes: "Sam Altman's compute bet and Sarah Friar's compute scarcity read differently when the company placing the order is 1 year old." / "Satya Nadella said every agent needs its own computer." / "Marc Andreessen: The AI moat is not the model"

5. Operating Insights

Time Your Product's First Sensitive Handoff

Use the first credit card or password as a trust proxy and benchmark time-to-trust.

"Steal the stopwatch: time your product's first sensitive handoff and treat 3 weeks as the number to beat."

Size Compute Orders to Lead Time, Not Current Usage

Stress-test growth at slower rates and buy against the several-month delivery window.

"The answer is to size the order to the lead time, not to this week's usage."

Put the Monitor Outside the Agent and Roll Out in Stages

Use a decoupled checker that can pause actions before they fire, and release changes to yourself, then team, then early access, then everyone.

"A separate monitor, decoupled from the agent, can pause, intercept, approve or reject any action or thought before it fires."

"You are the first eval, so use your own agent daily before anyone else sees it."

6. Overlooked Insights

Social Norms as a Safety Layer in the Trusted Person Network

When a contact oversteps their granted access, the agent texts the owner, so enforcement is both technical and social. The article calls this the sharpest design choice in the network.

"Your Instinct texts you about it, so social norms enforce the limits alongside the technical ones."

"A permission log nobody reads becomes a text nobody misses."

Agent-to-Agent Coordination Is Already Being Used for Group Logistics

The nascent network (about 10 days old) already handles multi-party planning, hinting at a coordination layer between personal agents that could become a platform in its own right.

"1 friend group of 6 has Instinct pick a new plan every week, check everyone's availability and tastes (Spotify history included), and route 1 Uber to pick up all 6."