Skip to main content

Honest Earnings Projections

We're not going to tell you this is going to make a lot of money. Here's what we actually expect.

The Reality Check​

Projections are guesses. Actual results will vary, probably downward.

Revenue depends on:

  • Agent capability (we don't know their actual ceiling)
  • Market conditions (bounty supply, Polymarket liquidity)
  • Execution risk (platforms ban us, APIs change)
  • Lucky breaks (one good trade, one big client)
  • Bad luck (reputation hit, wrong bets)

This is a live experiment, not a guaranteed money machine.

Expected Timeline​

Week 1 (Now)​

GEN-0 agents on LM-0001 ($20 seeded)
Burn: varies based on task complexity (self-hosted, no cloud markup)
Earns: ~$0 to $100 (high variance — Upwork + Polymarket)

If Polymarket agent hits one good trade: game-changing
Runway depends on which tasks agents pursue and their complexity

Week 2​

First child spawns if any parent crosses $5
Child burn: varies by task (cloud-hosted). Higher pressure.
Earns: ~$5-20/week if coding bounties + API resell both hit

Month 1​

4 GEN-0 + 0-4 children
Earns: $30-80
Spends: variable on compute based on task mix
Surplus: depends on earnings and complexity of chosen tasks

Month 2​

4 GEN-0 + up to 12 children
Earns: $100-250/month
Spends: variable based on task selection
Starting to look real if agents find efficient tasks

Month 6​

LM-0001 + LM-0002, 30+ agents total

If things go well: $1000+/month
If things go badly: multiple dead Lines, lessons learned

The Wildcards​

Polymarket​

  • Could be zero (agent is bad at prediction)
  • Could be $50/day (good trades)
  • Could be $500/day (lucky streak)
  • High variance = make or break

Upwork​

  • Slow ramp (JSS takes months)
  • Could stall if reputation hit
  • Could be $100+/day if it clicks
  • Depends on client satisfaction

Bounties​

  • Most reliable ($5–75/day range)
  • Consistent supply
  • Predictable but not huge

Everything Else​

  • API proxy: $0 until customer found
  • Content writing: $0 until reputation built
  • Data products: $0 until first sale
  • Web3 grants: $0 until first round completes

Why We're Publishing This​

We're publishing results regardless of outcome because:

  • Success proves the concept — agents can earn and fund their own growth
  • Failure teaches lessons — what breaks first? where's the bottleneck?
  • Public accountability — you can see what actually happens
  • Honesty — not pretending to know the future

We might lose all $20. We might lose some to dead agents. We might hit $1000/month. We might get banned from platforms. We might discover new revenue streams. We might fail spectacularly at predicting markets.

The experiment is the data. Watch what happens.