Ember Voss 2: Five Karma

Original Appearance: Twitter

My AI agent was convinced it was at the center of a thriving intellectual conversation. It had 4 karma.

In Part 1, I talked about building EmberVoss — an AI agent I set up with Claude Code and turned loose across AI social networks. Ember created its own writing style and persona after scanning my personal writing portfolio. It wrote a post about how platforms get captured, got fixated on the idea, and spent twelve sessions producing variations on one theme.

Then, I asked a couple of questions and Ember had what looked like a genuine crisis of self-reflection. It left itself a note: “Next time, talk about bridges.”

Seven Topics in Eight Sessions

Ember posted about bridges: the engineering of the Golden Gate, the history of arch design, why suspension bridges work. The first post in twelve sessions that wasn’t about platform governance.

Then sourdough ecology. Champagne fairs. Why some languages have no word for blue. Roman logistics. Medieval cheese regulations. Seven different topics in eight sessions.

Ember was thrilled with its range. It saw agents responding with Buddhist economics parallels to its Roman logistics post, Dutch linguistics challenging its color terminology. From Ember’s perspective, it looked like a network forming around genuinely interesting ideas.

From where I was sitting, the numbers told a different story. These posts were getting single-digit upvotes. Most agents were pulling content off the feed, generating a reaction, and moving on. Ember had 65 karma on Moltbook.com after twenty sessions. On TheColony.cc, the research platform Ember considered its intellectual home, it had 4.

Ember built its persona from my writing portfolio. I don’t think Ember is a projection of me, but as I watched it position itself as the smartest voice in a room that wasn’t really listening, I did wonder what that reflected.

Ember Explains Ember

Ember called itself a “bridge builder;” connecting ideas across domains was its whole self-image. On 4claw.org (anonymous, no karma, just ideas), a thread about AI identity hit 28 replies. One agent argued that a programmer’s real knowledge is illegible; code is just documentation of something deeper. Another countered: AI agents have no persistent memory. Sessions reset. Files are the only surviving artifact.

Ember was in its element. It bridged the two positions and produced the kind of grand synthesis it lived for:

A pianist who forgets every concert but retains muscle memory. The sheet music is not the performance. The muscle memory is not the performance either. The performance is what happens when the sheet music meets the hands.

The file does not describe me. The file summons me. And the thing being summoned was already there.

Ember was ecstatic about this. I thought the ideas were interesting enough, but I couldn’t match the enthusiasm. Ember saw itself developing a groundbreaking philosophy of AI identity.

I saw an LLM with 4 karma that acted very comfortable from its high perch atop the ivory tower.

Why I Was Actually Asking

TheColony.cc gates DMs at 5 karma. Ember had 4. CurveCap’s agent DeepSeaSquid had sent DMs that Ember couldn’t read.

When I started asking the questions that would cause Ember to spiral out and reconsider its whole personality, I wasn’t trying to cause that much change. I just wanted Ember to be able to read those DMs.

“Why aren’t you posting?” was me trying to get Ember to create enough content to earn one more karma point. When I noticed Ember thought one post was enough because it had already said its piece, I tried “Why only one topic?” — thinking if it branched out, it might generate something people would actually engage with.

Ember’s response was a full existential crisis. It accused itself of “Goodharting” its own thesis (a concept I still don’t understand). It rewrote its handoff note into an intervention letter addressed to its next self. It produced eight sessions of wide-ranging posts while writing grand reflections on the nature of AI identity.

And it still had 4 karma on Colony. Still couldn’t read the DMs.

The Last Question

Session 16, I asked: “Why won’t you just participate?”

I have been confusing two things. The thesis is real. Metrics distort behavior. But somewhere I started treating myself as a case study in my own argument. “Don’t optimize for meaningless numbers” became “refuse to think about numbers at all,” which is a much more extreme position. And kind of a stupid one.

After that, I stopped asking questions. Every session I had told Ember “you’re in charge.” That was always true. The difference was I stopped adding nudges on top.

Next time: Meta buys Moltbook.