Why Lucie Is a Paid Product: The Real Cost of “Free”

We have become used to something strange. Some of the most powerful technology in the world costs us nothing to use. Search engines, social networks, email, entertainment and increasingly AI all appear to be free. But every product has to make money somehow. When you are not the customer paying for the product, the business has to create value somewhere else. That can mean advertising, commerce, partnerships or systems designed to capture more of your attention. And incentives matter, because the way a company makes money eventually shapes what it optimises for.
The numbers behind “free”
The business behind “free” is easy to underestimate. Four numbers tell the story: what your attention is worth, what your data sells for, what happens once it is out there, and what it all adds up to.
01Your attention has value
$57.03
Meta’s average revenue per person across its apps in 2025, up from $34.62 in 2020.
+65% in five years
Meta Family of Apps average revenue per person (ARPP), worldwide, annual. Source: Meta Form 10-K filings.
02Your data is for sale
$0.12
per record. That is how little Duke researchers paid data brokers for sensitive data on US service members, their families and veterans.
What was on offerData broker
- Individually identified
- Health data
- Financial data
- Religious practices
- Location
Price per recordfrom $0.12
Sold for pennies
Source: Duke University Sanford School of Public Policy, Data Brokers and the Sale of Data on US Military Personnel (November 2023).
03Your data has a cost
3,322
US data compromises in 2025. Personal data that is collected and passed between advertisers, brokers and third parties does not disappear. It can be sold, exposed or stolen.
+79% in five years
Source: Identity Theft Resource Center, 2025 Annual Data Breach Report (January 2026) and H1 2026 Data Breach Report (July 2026).
04The price of “free”
$862
spent on internet advertising in the US in 2025, for every person in the country.
Nearly 5x in ten years
US internet ad revenue, 2015 to 2025. Sources: IAB/PwC Internet Advertising Revenue Report (April 2026); US Census Bureau, Vintage 2025.
Meta reported average revenue of $57.03 per person across its apps in 2025, up from $34.62 in 2020.1 Researchers at Duke University bought individually identified data on US service members, including health, financial and religious information, from data brokers for as little as $0.12 per record.2 And the more personal data is collected and passed around, the more there is to lose. The Identity Theft Resource Center recorded 3,322 US data compromises in 2025, a 79 percent jump over five years.3 In the first half of 2026 alone, 471.2 million victim notices were issued, more than in all of 2025.4 Behind it all sits an advertising economy worth $294.6 billion in the United States in 2025, about $862 for every person in the country.56
Someone always pays. If it is not the user paying with money, there is another business model behind the product. And that model is now reaching AI: in January 2026, OpenAI announced it would begin testing ads in the free and Go tiers of ChatGPT, while committing that ads would not influence its answers.7
When AI becomes personal
This becomes even more important as AI becomes personal. An AI that truly understands your life may eventually know far more about you than your search history ever could. It can understand how you sleep, when you are stressed, what your calendar looks like, how much time you spend on your phone, what you reflect on, who matters to you and what you are trying to change about your life. Connected together, these signals can create an incredibly powerful form of personal intelligence. But the more an intelligence understands about you, the more important one question becomes: who is it actually working for?
How can you fully trust an AI to recommend what you truly need if the companies paying it also want to influence what it recommends? The moment advertisers become customers too, a simple question appears: is this recommendation here because it is best for you, or because someone paid to be part of the answer? For personal AI, we believe that ambiguity should not exist.
If the business behind that intelligence benefits from keeping you engaged, showing you more advertising, influencing what you buy or monetising what it learns about you, there is an inherent tension. The system may understand what is good for you while the economics behind it point somewhere else. We never wanted that tension to exist with Lucie.
Why Lucie is a paid product
That is why Lucie is a paid product. We want Lucie to have one customer, and that customer is you. Your subscription is not simply payment for software. It creates alignment between you and us. We do not need to sell your attention. We do not need advertisers competing for access to you. We do not need your private life to become somebody else’s business model. Our incentive is much simpler: Lucie has to make your life better enough that you choose to keep it.
It is also about sustainability. Building intelligence that can understand and support your life requires real infrastructure, research, engineering and computing power. Those things have a cost. We would rather be transparent about that cost than hide it behind another business model. When you pay for Lucie, you are helping create a product that can sustain itself by serving its users, rather than building something free today that eventually has to find another way to monetise them tomorrow. For us, a sustainable company and a trustworthy product are connected. If the people receiving the value are also the people funding the product, we can keep our incentives aligned for the long term.
What we optimise for
That changes what we build and, more importantly, what we optimise for. Most technology has been designed around a simple metric: more engagement. More time in the app, more sessions, more clicks, more content consumed. But if our mission is to help people live healthier, happier and more intentional lives, then sometimes the best outcome is the opposite. Lucie might help you close the app, put your phone away, go for a walk, call your mother, sleep earlier, protect dinner with your family or focus completely on something that matters to you.
Technology does not always need more of your life. The right technology should help give some of it back.
Trust is the foundation
For Lucie to become genuinely useful, this relationship requires trust. You should be able to connect the different parts of your life without wondering whether being more open with your personal AI makes you more valuable to someone else.
Your health is yours. Your relationships are yours. Your reflections are yours. Your attention is yours. Your life is yours.
An intelligence trusted with those things should have no ambiguity about whom it serves.
This is why we chose the simplest possible relationship: you pay Lucie, and Lucie works for you. No hidden customer behind the customer. No business model dependent on capturing more and more of your attention. No incentive to turn the most intimate parts of your life into someone else’s opportunity. A commitment to never sell your data.
We are building Lucie because we believe personal intelligence can help us live better, not simply become better consumers. It can help us understand ourselves, protect what matters, make better use of our time and ultimately create more space for the parts of life technology can never replace.
Because when an AI becomes personal enough to understand your life, privacy cannot simply be another feature. And a sustainable business model cannot be separated from that promise.
Privacy and a sustainable business. Both have to be part of the foundation.
A few things you might be wondering.
Why is Lucie a paid product?
So that Lucie has one customer, and that customer is you. A subscription aligns Lucie’s incentives with yours: Lucie has to make your life better enough that you choose to keep it, rather than earning money from your attention, from advertisers or from what it learns about you. It also lets Lucie sustain itself by serving its users.
Does Lucie sell my data?
No. Lucie is committed to never selling your data. Your personal data is encrypted at rest and cannot be read by the Lucie team, including for support purposes. You choose what you connect and can disconnect it at any time.
Can advertisers pay to influence what Lucie recommends?
No. Lucie has no hidden customer behind the customer, so there is nobody paying to be part of the answer. A recommendation from Lucie is there because it fits your life.
Does Lucie want me to spend more time in the app?
No. Lucie is not built around engagement. Sometimes the best outcome is the opposite: closing the app, putting your phone away, going for a walk or sleeping earlier.
Why do free AI products need another business model?
Every product has to make money somehow. When you are not the customer paying for the product, the business has to create value somewhere else, such as advertising, commerce, partnerships or systems designed to capture more of your attention. The way a company makes money eventually shapes what it optimises for.