The Technological Covenant: 22 Principles We Build By

Silicon Valley owes a moral debt to the families whose lives it has rewired. The engineering elite has an affirmative obligation to the users it has made dependent on its products — beginning with the children who never consented to be the raw material.

We must rebel against the tyranny of the feed. Infinite scroll has changed our lives, but its genius is our captivity. Attention taken from a parent is presence taken from a child; what the feed gains, the dinner table loses.

"Free" is the most expensive word in software. A product is not free when the price is paid in attention, in data, and in the slow erosion of a parent's ability to be present with their own kid. The bill always comes due, usually to the next generation.

The limits of surveillance capitalism have been exposed. The flourishing of free societies in this century will not be built on software that watches its users. It will be built on software that serves them — and answers to them.

The question is not whether AI will be pointed at children; it is who will point it and for what purpose. The engagement optimisers, the data brokers, and the attention merchants will not pause to indulge in theatrical debates about youth mental health. They will proceed. So must we — in the opposite direction.

Protecting children online should be a universal duty. We should, as a society, seriously consider whether anyone profiting from a product used by a child has met the obligation that profit entails, or has simply externalised the cost onto the child's nervous system.

If a parent asks for a better tool to stay connected to their kid, we should build it. We should remain capable of debating the role of technology in family life while remaining unflinching in our commitment to the parents we have asked to raise children in a world we built and then walked away from.

Platform operators need not be priests — but they should not be wolves. Any business that compensated itself the way engagement platforms compensate themselves — in increments of their users' attention and their users' children — would, in any other industry, be regulated out of existence.

We should show far more grace toward parents navigating a world the tech industry made harder. The eradication of any space for imperfect parenting — any tolerance for the ordinary struggles of raising a child — leaves families with a cast of algorithms and influencers they will grow to regret listening to.

The psychologisation of modern parenting is leading us astray. Those who look to the feed to nourish their sense of self as a parent, who rely on performing their family life to audiences they will never meet, will be left disappointed — and so will their kids.

Our industry has grown too eager to celebrate the demise of attention, privacy, and unmediated childhood. The vanquishing of these is not a milestone to cheer. It is a moment to pause — and, if we have any honesty left, to reverse.

The age of the ad-funded platform is ending. One era, built on the premise that humans are the product, is exhausting its social licence. A new era — built on the radical idea that the customer is the customer — is set to begin.

No other medium in history has done more to connect families across distance than the modern internet. The internet is far from perfect. But it is easy to forget how much more a grandparent can see of a grandchild's life today than in any prior era — and easy to forget that this connection is worth protecting from the people who would monetise it to death.

Strong privacy norms have made possible extraordinarily safe childhoods for those who still have them. Too many take for granted that a child's life used to unfold in relative obscurity — that three generations ago, a child's awkward moment died in the room it happened in, not on a server somewhere, forever.

The postwar dismantling of consumer protection must be reversed. The defanging of privacy law in the United States was an overcorrection for which families are now paying a heavy price. A similar commitment to letting platforms self-regulate will continue to shift the balance of power between platforms and the people whose lives they govern.

We should applaud those who build where the market has ignored the quiet, unscalable work. The culture sniggers at the founder who serves a small, specific audience deeply — as if software ought to optimise for reach rather than service. Any curiosity about software that chooses depth over scale is dismissed, or lurks beneath a thinly veiled scorn.

Silicon Valley must play a role in addressing the crises of attention, loneliness, and adolescent mental health it has been central in producing. Many founders have essentially shrugged, abandoning serious effort to address problems their products helped create, in what should be a desperate bid to save childhoods.

The ruthless exposure of ordinary private lives — especially children's — drives far too much of what ails us. The archive has become so unforgiving that a generation of kids is left choosing between disappearing entirely and performing a version of themselves for strangers.

The caution that surveillance teaches children is corrosive. Those who grow up knowing every word is logged, every photo permanent, every friendship documented, learn to say nothing wrong — which, in the end, means they say nothing much at all.

The pervasive intolerance in certain circles for slow, offline, unrecorded family life must be resisted. The industry's disdain for lives that do not produce content is one of the most telling signs that its project is a less humane movement than many within it would claim.

Some business models have produced wonders for their users; others have proven extractive and corrosive. We are told all models are now equal — SaaS, ad-funded, data-brokered, "free." Yet this dogma glosses over the fact that certain models — where the customer pays and the company serves — have built generations of trusted tools. Others have produced an industry whose leaders do not let their own children use the products they built.

We must resist the shallow temptation of a hollow neutrality. The idea that every technology is just a tool, every platform just a venue, every business model just a preference. We have, in software, for a generation refused to take a position on what technology is for. But if it is not for the flourishing of the person using it, then what, exactly, is it for?