There's a better way, and it became possible only recently: the AI assistant you already use is extremely good at reading bank statements.

The idea

Any bank in the world can export a statement — PDF, CSV, sometimes Excel. Those files are messy and every bank formats them differently, which is exactly the kind of unstructured-to-structured problem modern AI models solve in seconds. So instead of building fragile bank integrations, Gilt does something simpler:

• Gilt gives you a prepared prompt that describes precisely the file format it can verify and import — columns, dates, currencies, signs.
• You hand that prompt, together with your statement, to your own AI — Claude, ChatGPT, whichever assistant you already have, in their app, under your account.
• The AI returns a structured file. Gilt validates every row — dates, amounts, duplicates, balances — and shows you the result for review before a single transaction is saved.

Why this is more private than “connecting your bank”

Bank-linking flows route your data through an aggregator's infrastructure and often require your actual banking credentials. With the bring-your-own-AI approach, exactly two parties see your statement: your bank, which already has it, and the AI provider you chose and already trust with your questions. Gilt's own app makes zero network calls to any AI service — it just verifies a file that never leaves your phone.

Why it's more universal

Aggregators cover a few thousand banks, mostly in the US and Western Europe. Statements exist everywhere. A small bank in Armenia, a neobank in Brazil, a credit union with a website from 2009 — if it can produce a PDF, this works. That's not a corner case; for most of the world, it's the only way any budget app will ever import their data.

What review-before-save looks like

AI models are good, but “good” isn't “trusted with your ledger.” That's why the import screen shows every parsed transaction — amount, date, category suggestion — and nothing is written until you confirm. Bad row? Fix or skip it. The AI does the tedious part; you keep the authority.

Six statements, six accounts, one evening

In practice, importing a year of history across several accounts takes an evening of coffee-length sessions: export statement, paste prompt, drop the file back into Gilt, review, done. Compare that to retyping a thousand transactions — or to giving a data broker your bank password.

Two ways to build a finance app

The standard way: your transactions sync to the company's backend, where they power features, analytics — and, at many free apps, monetization. You are trusting an unending series of decisions by people you'll never meet: every engineer with production access, every acquisition, every “update to our privacy policy” email.

The other way: the app stores data on your device and syncs it through your own cloud account — for iPhone, that's your personal iCloud with its CloudKit private database. The developer operates nothing. There's no admin panel where your spending exists, no database to breach, nothing to subpoena, nothing to sell.

“We can't” beats “we won't”

• Breaches: you can't leak a database you don't have. Your data's security becomes Apple's iCloud security — audited at a scale no indie backend can match.
• Business model drift: a company that never sees your data can't be tempted to monetize it later.
• Longevity: if the developer vanished tomorrow, a serverless app keeps working. Your ledger doesn't die with a startup.

How to check any app in 30 seconds

Open its App Store page and scroll to the privacy label. “Data Not Collected” is rare in finance — most trackers collect identifiers, usage data, sometimes financial info itself. Then ask one question the marketing page won't answer: does this app work without creating an account? If it demands an email, there's a server behind it, and your data is on it.

Gilt's label reads Data Not Collected, and there is no account to create. Not because we promise not to look — because we built it so there's nowhere to look.

What a family budget actually needs

• One ledger, two phones. Both partners see the same accounts, transactions and budgets, live.
• Separate identities. Each person keeps their own Apple ID, their own Face ID, their own device security.
• Both can write. A budget only works if whoever bought the groceries logs the groceries.
• One payment. Nobody wants to buy the same app twice.

How Gilt does it

Two Apple technologies do the heavy lifting. iCloud sharing (CloudKit) carries the data: you invite your partner from the app, and the household ledger syncs directly between your personal iCloud accounts — no third-party server in between, ever. Apple Family Sharing carries the subscription: one Family plan purchase unlocks the app for up to five members of your Apple family group automatically, on their own devices.

The result feels almost boring, in the best way: your partner opens the app on their iPhone, everything is just there, and it stays in sync. No invite codes to a company's cloud, no shared login to rotate when a phone gets lost.

Multi-currency households

Plenty of families earn in one currency and spend in another — or split life between countries. A shared budget has to handle transfers between currencies without inventing money out of exchange-rate rounding. Gilt treats multi-currency as a first-class case: accounts keep their own currencies, transfers link both sides, and reports convert honestly.

And the privacy part

A household's finances are among the most sensitive data two people share. Everything above happens inside Apple's ecosystem — your devices, your iCloud accounts. Gilt has no servers and cannot see a single shared transaction. That's by architecture, not by policy.