We're Not Building a Chatbot. We're Building the Half Hour Before You Go Live.

Every few months someone asks us, half-joking, whether ETAPX is a live-streaming company or something bigger wearing a live-streaming company's clothes. We've usually dodged the question, because the honest answer used to sound like spin: yes, and. Now there's a less abstract reason to answer it straight. We're planning a new AI app, built as its own standalone product, and it's early enough that we're not going to pretend otherwise. There's no ship date to announce here, and no feature list worth printing yet. What we can do is explain the reasoning, because the reasoning is the actual news, and because we'd rather say the honest, unfinished version of this now than a polished version once it's too late to be useful to anyone deciding whether to pay attention.

Why a live-streaming company would build an AI app

The short version: ETAPX was never really a live-streaming company. It's a company that keeps building tools for the specific moment when someone tries to turn attention into something real, whether that's income, a physical product, or a smaller, more honest audience than a public feed gives them. Whistlr happens to be the product where that thesis is easiest to see, because the mechanics are visible in real time. A creator goes live with no follower minimum, from a native Studio or any RTMP encoder, viewers send Gems as direct support, and the money lands in a day or two instead of sitting in a thirty-day hold somewhere. That loop, attention converting into something tangible, fast, without gatekeeping in front of it, is the pattern we keep building toward. It is not the exception inside ETAPX. It is closer to the whole point.

An AI app is not a departure from that pattern, even though on paper it looks like the furthest thing from a live-streaming feature. It's an attempt to attack a different part of the same problem from the other end. Whistlr solves the distribution and payout side of turning attention into outcomes, and it solves it well. It does nothing, on its own, for the thinking that happens before someone ever goes live, before they price a Storefront listing, before they decide what belongs on Circuits versus what gets saved for Flow. That thinking is still manual, still scattered across notes apps and half-remembered instincts, for almost everyone building something online, creator or not. We think there's a real product in closing that gap, and we don't think it belongs bolted onto one app as a feature nobody asked for and half of them will ignore.

The problem it's actually aimed at

Here's the pattern we keep running into, across our own team and every creator conversation that goes deep enough to be useful. The constraint on most people building an audience or a small business isn't willpower and it isn't talent, and it's rarely a missing tool either. It's decision fatigue spread across too many disconnected surfaces, each one demanding a small choice that feels too minor to plan for and too consequential to skip. What to stream tonight, and for how long. What to price a Gem tier at on the WTC tab. Whether a product belongs in PX Shop or a creator's own Storefront. Whether a post should go to a close circle on Circuits or the open feed on Flow. Each decision is small on its own. The volume of them, repeated daily, compounding across weeks, is what actually breaks people's momentum long before burnout gets the credit for it.

That's the shape of the problem this app is aimed at. Not another feed, not another chat window bolted onto an existing product for the sake of having AI somewhere in the changelog, and not a generic assistant that answers trivia and writes captions on request. Something closer to a working layer that sits alongside the decisions people are already making, across whatever tools they're already using, and makes the next one faster and better-informed than it would have been on instinct alone. If it works, most of the value should be invisible: fewer stalled plans, fewer mispriced offerings, fewer nights where someone meant to go live and didn't because they hadn't decided what to say.

Where it fits next to Whistlr, not inside it

We're building this as a standalone product on purpose, and it's worth explaining why, because the instinct at most companies our size is to jam a new capability into the app that already has the users and call it a feature launch. ETAPX runs more than one product for a reason that has nothing to do with empire-building. Whistlr is the live and creator-monetization surface. Influxx, Ocsidian, and GLSRM each exist because a single app was the wrong shape for the problem they were built to solve, and forcing them into one codebase would have made all of them worse. An AI app that only worked inside Whistlr would inherit Whistlr's shape and its assumptions, and the thinking we're describing doesn't only happen while someone's live or checking their Earnings tab. It happens before any of that, and it happens for people who may never open Whistlr at all.

That said, the connection to what already exists isn't incidental, and we're not pretending it is. A creator managing Gems, a WTC tab full of priced offerings, a Storefront, and the split between Circuits and Flow is already juggling more surfaces than most small businesses ever have to handle at once. If this app does its job, the clearest early proof of it should show up exactly there: a Whistlr creator planning a week of streams, thinking through pricing on WTC before publishing it, or deciding what belongs in PX Shop versus their own storefront, doing that thinking somewhere purpose-built instead of in a stray notes file or a text thread to themselves. Standalone doesn't mean unrelated. It means the tool has to earn its place with anyone who has this problem, not just people already inside one ecosystem, before it gets to lean on being convenient for existing users.

• The gap between deciding to create something and actually starting, which is where most momentum quietly dies before anyone notices it's gone
• The scattered decision-making that happens across separate apps, tabs, and half-finished notes before a creator ever goes live or lists a product
• A tool that works whether or not someone uses any other ETAPX product, because the problem it solves isn't specific to Whistlr or to creators at all
• A companion to the existing ecosystem rather than a replacement for any piece of it, including Whistlr's live tools, Storefront, or Circuits and Flow

The tradeoff we'd rather name than hide

Here's the part most teaser posts skip, and it's worth saying plainly instead of behind a wall of optimism. Building a standalone AI product instead of a Whistlr feature is slower, harder to market, and easy to misread as a distraction from the app that already has an audience. A feature ships behind an existing login and inherits existing trust. A standalone product has to justify itself from a blank screen, with nobody obligated to open it. We're choosing the harder path on purpose, because bolting this onto Whistlr would solve our distribution problem and completely dodge the actual product problem, which is that the thinking that matters most happens outside any one app's walls. Optimizing for an easy launch and optimizing for the right shape of product turned out to be different decisions, and we picked the second one.

What we're not saying yet, and why that's deliberate

We're not going to describe specific features here, and we're not going to pretend we have more certainty than we do. We don't have a launch window worth printing, and anyone who tells you otherwise about a product still in active shaping is selling you something. What we do have is conviction about the problem, real internal prototyping, and enough of it to know the shape is right even though the edges are still moving week to week. That's a teaser, not an announcement, and we'd rather be honest about which one this is than dress it up as more finished than it is.

We keep asking the same question about every ETAPX product before it ships: does this remove a real decision someone was stuck on, or does it just give them one more thing to check. If we can't answer that plainly, it's not ready.

ETAPX, internal product principle

Why this is worth the risk of saying anything at all

Standalone AI products are a crowded, noisy category right now, and we're aware that announcing intent this early invites comparisons we can't yet live up to. We're doing it anyway, because AJ has been consistent about one thing since the earliest days of Whistlr: build for the moment before the moment everyone else is optimizing for. Every other player in the creator economy is fighting over the live stream itself, the feed itself, the payout itself, all the visible, demo-able moments. Almost nobody is building for the half hour before any of that happens, when a person is just sitting there trying to figure out what to do next. That's an unfashionable place to build a product, because it doesn't screenshot well. It's also, we think, an underbuilt one, which is exactly the kind of gap ETAPX has tended to notice before it gets crowded.

So here's the honest framing, plainly stated instead of buried under a features list. ETAPX is planning a new AI app because the same thesis behind Whistlr, Influxx, Ocsidian, and GLSRM points there next: find the moment where someone's stuck turning effort into something real, and build the tool that unsticks it, whether or not that moment happens to live inside an app we already built. We don't have a ship date, and we're not going to invent one just to make this post feel more finished. What we have is a direction, worked on with the same seriousness we put into Gems payouts landing in a day instead of thirty, and we'd rather tell you where we're pointed before we're finished than let you find out after the fact.