Newly’s $2 Million Raise Now Runs Into Apple’s Review Wall

Stockholm’s Newly has raised more than $2 million for an AI agent that writes iOS and Android apps and tries to walk them through store review. The company, formerly Natively, says the grind is no longer writing the binary. It is getting Apple and Google to let that binary live.

Founder and CEO Timothy Lindblom has sold the shift as shrinking build time “from months to hours.” The March 23, 2026 round, led by PSV Tech, is the cash behind that claim. The product the money is actually buying is a reviewer simulator, and Apple spent June rewriting the rule that simulator now has to beat.

Newly Raised More Than $2 Million to Clear Review

Newly’s own site puts the total at more than $2 million from builders and operators. PSV Tech led, with Karaoke Club, Wave Ventures, Inception Fund, Foundry Ventures, Tiny Supercomputing Investment Company, and a cluster of Nordic angels. The firm was founded in 2025 and is based in Stockholm.

On the same day as the raise, Lindblom rebranded Natively to Newly and posted the launch as an agentic app builder. The clip is still the cleanest public demo of what the company thinks it is selling.

Introducing Newly.

The world's first Agentic app builder.

RT + Comment "Newly" and I'll send you 100 apps to make $10K/mo. pic.twitter.com/Dt7PONfnQd

— Timothy Lindblom (@timothylindblom) March 23, 2026

The wire version of that day talked about no-code native apps. Lindblom’s own line was narrower. People can already spin up full mobile apps fast, he has said. The blocker is the App Store.

FROM PROMPT TO STORE, 2025 TO 2026

1. 2025: Natively is founded in Sweden and starts shipping an AI builder that outputs mobile apps.
2. March 23, 2026: The company rebrands to Newly, announces more than $2 million in funding, and pitches an agent that aims at store-ready builds.
3. June 8, 2026: Apple updates the App Store Review Guidelines, including a harder spam rule that can also pull live apps.
4. June 2026: Newly publishes a how-to on submitting an AI-built app, arguing the binary must be native and filed by the content owner.
5. September 10, 2026: Lindblom is still telling followers the stack is Expo and React Native.

That calendar is the whole bet. Generation got cheap first. Review did not.

The Product Is a Review Agent

Newly describes itself as the world’s first end-to-end compliance-aware app agent. The about page is blunt: building the app is not the hard part, getting it approved is. The agent is supposed to check store rules while you prompt, prep screenshots, privacy policies, terms, and metadata, and steer the file away from a bounce.

The output is not Swift written by hand. Apps come out as React Native on Expo, which Newly says compiles to real iOS and Android binaries rather than a WebView shell. Buyers get the source as a ZIP or a GitHub repo. A bundled service, Liquid Backend, handles database, login, files, websockets, and functions inside the monthly fee. Mail goes out through Resend. SMS goes through Twilio. Publishing is pointed at Expo Launch.

WHAT THE AGENT SAYS IT HANDLES

• Rule checks: It flags App Store and Play Store issues while the app is still being prompted, not after a rejection email.
• Listing pack: It drafts screenshots, privacy policy, terms of service, and store metadata with the build.
• Native binary: It targets a compiled React Native and Expo app, which Newly contrasts with a website wrapped in a shell.
• Your account: You submit under your own developer account, which is the path Apple’s template rule allows.

Apple’s Guideline 4.2.6 rejects apps from a commercialized template or app generation service unless the provider of the content files them. Newly’s how-to, updated in June 2026, hangs the whole product on that clause. The AI is not the obstacle, the page says. A thin website pretending to be an app is. So is a factory of identical clones filed from the builder’s own account.

Apple still charges $99 a year for the Developer Program. Google Play is a one-time $25. Those fees sit on top of Newly’s own bill, and no vendor can promise a pass. Newly says that in public, which is the honest part of the pitch.

Apple’s June Rule Hits Apps Already on the Store

Apple already rejected a lot of uploads before anyone called this vibe coding. Its 2025 App Store transparency report is the latest full-year count, and it is the floor under Newly’s promise.

APP STORE REVIEW AT APPLE, 2024 AND 2025

The share of uploads that fail is slightly down, 24.9% to 23.0%. The pile is not. Apple reviewed about 1.33 million more submissions in 2025 than in 2024, and it took down about twice as many live apps, 166,899 against 82,509. Another 387,087 submissions got through in 2025 only after a first rejection, which is the working definition of review as a loop, not a door.

On June 8, 2026, Apple refreshed the App Store Review Guidelines. The spam clause, 4.3(b), is the one that now sits on top of a tool like Newly. It no longer only blocks a new fart app at the door. It tells developers not to file apps that are indistinguishable from what is already widely available, and it says opportunistic variants of existing categories degrade discovery. Dating, flashlight, sound effects, wallpaper, simple timers, and fortune telling need a meaningfully different experience. Apple says it may remove apps in those lanes if they are not updated, improved, or drawing customers. Drinking games, Kama Sutra, fart, and burp apps are called out as low-effort, and repeat filers can lose the Developer Program.

Don’t submit apps that are indistinguishable from what’s already widely available. Opportunistically creating variants of existing app categories or popular apps degrades App Store discovery, reduces overall app quality, and harms both users and developers.

App Store Review Guidelines, Guideline 4.3(b), June 8, 2026

Guideline 4.2 still dumps a “repackaged website.” That is the WebView problem Newly designed around. 4.3(b) is a different test. It asks whether the app is worth another slot, not whether the code is native. A clean Expo binary can still look like ten thousand other prompt-built utilities.

The Playbook That Copies What Already Works

Lindblom has not only been selling a compliance agent. In the weeks around the raise he posted a growth script for first-time builders that starts with someone else’s hit. Pick an app that crossed 10K downloads fast, scrape the one-star reviews, paste the complaints into a model, and ship a tighter version. He called it the easy way to build a viral app that makes money. A later thread told non-technical founders to steal the blueprint of apps already minting cash rather than invent a category.

That advice is rational if review is a paperwork problem. It is radioactive if review is a sameness problem. 4.3(b) is now written against opportunistic variants. A feed of nearly identical weather skins, AI wrappers, and “fixed what you hate” clones is exactly the inventory Apple says it will starve. The louder issue around vibe-coded mobile apps is not that the compiler fails. It is that the winning tactic on X is to photocopy a chart-topper and hope the listing looks original enough.

Newly’s how-to tries to split that hair. A unique native app you own and file yourself is allowed, it says. A commercialized template clone is not. The gap is taste, and taste is what human reviewers, plus Apple’s own screening models, get to define. Lindblom cannot prompt his way out of that, and neither can a $25 subscriber.

King, Stardoll, and a Copenhagen Seed Check

The cap table is more revealing than the blog posts. These backers have already lived inside hit mobile products, which is why they can price a tool that multiplies store inventory.

WHO WROTE THE CHECKS

• PSV Tech: Danish lead investor, now raising PSV Tech Fund II with a €70 million target for Nordic AI, software, and infrastructure bets.
• Sebastian Knutsson: King co-founder, Candy Crush veteran, and an LP in that PSV vehicle, now an angel on Newly.
• Mattias Miksche: Stardoll, the old social-doll network that taught a generation of Swedes how consumer apps scale.
• Peter Carlsson: Northvolt. A hardware name on a software round, sitting beside Google alum Joseph Michael and Grafbase’s Fredrik Björk.
• Karaoke Club and Wave Ventures: A Copenhagen builder fund founded in 2025, and a Helsinki student-run firm that closed an $11 million Fund III in June 2026.

Knutsson has said the Nordics have the safety net and the schools, and that founders still leave for U.S. capital. Newly is the opposite trade: keep the builder in Stockholm, sell a global App Store slot as a subscription. The people who extracted years of specialised labour from native clients are now funding a machine that makes that labour look optional. If the agent works, the loser is the small iOS shop that used to quote in months. If 4.3(b) holds, the loser is the subscriber who ships a lookalike and burns a year of Apple fees on a listing that never sticks.

What $25 a Month Buys

Newly’s public floor is simple, and it is cheaper than a contractor week.

NEWLY’S LIVE PLANS

Credits get burned as you prompt, so a heavy week costs extra or a higher tier. Annual billing is advertised as a discount on the live cards. Discord is over 1,000 members. The interesting line is Apex. The features that most closely match the “compliance-aware” pitch, automated store checks and listing work, still sit on a coming-soon card. The seed round is already in the bank. The product that most needs to beat June’s rule is the one that is not for sale yet.

Even a clean export does not retire the stores. Apple is $99 a year. Google is $25 once. Review can still send the build back for crashes, missing privacy copy, a thin feature set, or a listing that looks like a neighbour. Newly says you own the code forever and can leave. That is a real difference from older no-code runtimes that trapped the project inside their own player. It does not buy you a search ranking.

Seed Money Still Has to Get Past Cupertino

The March plan also talked about leaving the phone. Desktop software, augmented reality, and wearables were on the use-of-proceeds list. Those surfaces have their own gates, and none of them are cheaper than App Review. With Apex still dark and the live SKU sitting at $25, the near work is still iOS and Android.

Lindblom’s September 10 note was a small one, a reply that Newly is on Expo and React Native, which is the same stack the docs already sell. Six months after the raise, the company has not posted a new round, a public approval rate, or a count of apps that made it through Apple on the first try. What it has is a reviewer-shaped agent, a Nordic cap table that knows what a mobile hit looks like, and a store that spent June saying it will delete the boring copies.

The next test is dull and specific: whether a prompted Expo app, filed by its owner, with a privacy policy the agent drafted, survives 4.2 and 4.3 in the same week. Until Apex ships, that test runs on the $25 plan, one credit stack at a time.