AI on social media has become less a feature than a foregone conclusion: chatbots in the inbox, AI companions in the feed, algorithmically generated posts filling the space between real ones. Whistlr doesn't do any of that, on purpose. We keep the feed built for authentic human connection genuinely human, and channel everything ETAPX builds around artificial intelligence into GLSRM — a completely separate initiative built for exactly that instead.
The question of Whistlr and AI comes up often enough to deserve a direct answer. Most social platforms treat adding an AI layer as an easy call: ship a companion, let a model fill the feed when real posts run thin. We looked at that roadmap and chose the opposite — not because artificial intelligence is unimportant, but because it's important enough to deserve its own product, built by people whose entire job is getting it right.
The Industry Trend We Looked At and Said No To
Look at any major social platform's roadmap and you'll find some version of the same three features: an AI chatbot in the inbox, an AI companion built to keep someone talking when no one else is around, and a feed quietly padded with AI-generated posts to keep pace with demand. It's become the default assumption of the industry — an algorithmic feed needs an AI layer the way it needs a like button. We evaluated whether Whistlr should have some version of that. Capability was never the constraint; GLSRM proves ETAPX has the expertise. The real question was whether it would make Whistlr better at the one thing it exists to do: connect real people to each other. The answer was no.
Why Whistlr Doesn't Bolt AI Onto the Feed
Whistlr's entire premise rests on one promise: the person on the other side of a message, a livestream, or a post is a real human being, not a simulation of one. Creator monetization, Circuits, Waves, live streaming — everything is built downstream of that promise. The moment a user has to wonder whether a comment came from a bot, or whether a companion they've messaged for weeks is a person or a feature, the promise weakens. A platform that markets itself on authentic connection can't afford a diluted version of the one thing it sells.
This isn't a claim that AI companions or AI-generated feeds are bad ideas — plenty of products are built around exactly that premise, openly, and their users know what they're getting. The problem is narrower: those features don't belong inside a product whose entire pitch is the opposite. Bolting an AI companion onto a platform built for authentic human connection doesn't add a feature — it quietly changes what the platform is for. The result, on purpose, is close to an AI-free social platform where it matters most: the feed, the messages, the streams.
A few specific tradeoffs kept coming up in that debate, worth stating plainly.
- Authenticity is binary, not partial: once a user can't be certain a comment is human, they extend that uncertainty to the whole feed, real parts included.
- Creator trust is the business model: monetization depends on supporters believing a creator relationship is real; an AI layer nearby risks that exchange.
- Engagement-optimized AI has different incentives than connection: a companion built to maximize time spent talking to it works against, not for, two people building something real.
- AI-generated filler tends to look the same everywhere: synthetic content converges toward one generic style — the opposite of a distinct, creator-first feed.
- Half-measures rarely stay half-measures: a small AI feature tends to expand once it exists, and the safest move is to not open that door.
"We could have added an AI companion to Whistlr in a quarter. That was never the hard part. The hard part was admitting it would make the feed worse at the one job it has, and choosing not to ship it anyway."
— Nadia Kessler, Head of Product at ETAPX
Where We Think AI on Social Media Goes Wrong
The broader debate about AI on social media tends to get flattened into a binary: either AI belongs in the feed, or you're against AI altogether. We don't think that holds up. Feeds built around broadcast or discovery, where users consume content from strangers, can absorb an AI layer without breaking their own premise, because the premise was never "you are talking to a specific, known person." Whistlr is built differently: knowing the person is real is the point. The same feature — an AI companion, an AI-generated post, a filler feed — can fit one platform and not another, without either being wrong about AI in general. Context decides the answer, not a blanket rule.
Why We Built GLSRM Instead of Adding AI to Whistlr
If we thought AI mattered less than the rest of the industry seems to believe, this decision would be simple. We don't. GLSRM exists because we think the opposite: AI is consequential enough to deserve the dedicated coverage finance has had with Bloomberg, and sports and mainstream tech have had for decades. GLSRM was born out of ETAPX's own frustration trying to keep up with AI industry news through a dozen scattered sources, even as a team that builds AI-adjacent products for a living. If we couldn't keep up with the tools that already existed, almost no one else could either.
That's a real editorial mission, not a feature bolted onto an app. GLSRM is built around actual editorial standards rather than aggregation, with the goal of becoming the front page of AI the way a market open is the front page of finance. Once we'd committed to that seriousness somewhere, the case for stuffing an AI companion into Whistlr's inbox got weaker, not stronger. Splitting the two wasn't a compromise — it's how each gets to be good at its job.
Two Products, Built for Two Different Sets of Values
Whistlr and GLSRM are both ETAPX products, shaped by artificial intelligence mattering to us — just in opposite ways. Whistlr is optimized for authentic human connection: real people, real relationships, real creators, across the general feed, Waves, and Circuits. GLSRM is optimized for rigorous AI journalism: sourcing discipline and the editorial judgment behind AI Race Insights, our piece on evaluating AI benchmark claims without getting misled by a leaderboard. That split is the ETAPX AI strategy in one sentence: two products serving two different sets of values instead of compromising both inside one. A social feed asked to double as a serious AI product gets worse at connection; a journalism product asked to double as a friendly feed gets worse at rigor.
This is consistent with how we approach anything sensitive across ETAPX. We've written before about keeping Whistlr safe without making safety loud — closed-loop sharing, moderation, and identity verification work quietly in the background rather than as heavy-handed or gimmicky features. The same instinct applies here. The right answer to a hard product question is rarely the loudest one, and in this case the quiet answer was to build GLSRM instead of a chatbot.
"The quiet answer almost never gets a headline. Not building the AI companion doesn't make a keynote slide. But it's the same instinct that keeps safety working quietly in the background instead of as a badge: do the considered thing, not the loud one."
— Nadia Kessler, Head of Product at ETAPX
A Deliberate Bet, Not a Verdict on AI
We want to be direct about what this is and isn't. This is a deliberate, sometimes contrarian bet against a real industry trend — most platforms are adding AI layers to their feeds, and we're choosing not to, knowing the choice looks conservative to some and out of step to others. It isn't a claim that AI is bad, or that AI companions and AI-generated content are illegitimate everywhere. GLSRM is proof of the opposite: we think AI is important enough to build an entire initiative around covering it seriously.
It's a narrower claim, and a more useful one: AI doesn't belong bolted onto a social feed whose entire premise is real people connecting with other real people. That's not a hedge, and it isn't a permanent statement about the industry at large. It's the reasoning behind one specific decision, stated plainly rather than dressed up as something bigger.
"Every other app I use has started talking back to me. Whistlr is the one place left where I know a comment came from an actual person, and that's exactly why I still trust what I read there."
— Reece Talvera, ETAPX community member
Frequently Asked Questions
Does Whistlr use AI-generated content in its feeds?
No. Whistlr's feeds — the general feed, Waves, Circuits, and the rest — are built around content from real people: creators, friends, and communities. Filling that with AI-generated posts or video would undercut the authentic connection the platform is built around.
Why is there no AI feed on Whistlr?
Because an AI-populated feed — AI companions in the inbox, AI-generated posts in the scroll — changes what users are looking at without them realizing it. Whistlr's feed is built around real people on purpose, so an AI feed would work against its own premise.
Is ETAPX against artificial intelligence?
No. ETAPX built a separate initiative, GLSRM, specifically because AI is significant enough to deserve dedicated, serious coverage. The position isn't that AI is bad — it's that AI features don't belong bolted onto a feed built around authentic human connection.
What is GLSRM and how is it related to Whistlr?
GLSRM is a separate ETAPX initiative focused on dedicated AI industry news, described internally as the front page of AI. It shares a parent company with Whistlr but not a product surface, built so ETAPX's AI ambitions could be pursued with real editorial standards.
What is ETAPX's AI strategy across Whistlr and GLSRM?
Keep AI out of Whistlr's core feed so authentic human connection stays uncompromised, and channel ETAPX's AI ambition into GLSRM instead, where it can be built around rigorous editorial standards rather than diluted into a consumer social feature.
Could Whistlr ever add AI features in the future?
Never say never about any roadmap, but this isn't a temporary limitation — it's a considered position about what Whistlr is for. Any future feature would have to clear the same bar this one did: does it strengthen authentic connection, or work against it.
None of this predicts where the rest of the industry ends up, and it isn't a judgment on platforms that choose differently. It's simply ours: keep Whistlr's feed built for real people, and let GLSRM carry ETAPX's AI ambitions somewhere they can be pursued properly. Two products, two jobs, no compromise on either.
