Twin AI raises $10M seed to build agentic computer-use platform, claims 100K+ agents deployed
Jan 28, 2026
Key Points
- Twin, a Paris-based AI agent platform, raises $10M seed funding with over 100,000 agents deployed, positioning itself as a proprietary alternative to open-source competitor Moltbot.
- Moltbot's open-source model running locally on user devices sidesteps licensing and liability constraints that bind enterprise AI labs, explaining its rapid adoption despite similar technical capabilities elsewhere.
- Twin's institutional backing signals investor confidence in agentic platforms as a durable category, though whether a proprietary approach can compete with Moltbot's structural advantages remains unresolved.
Summary
Twin, a Paris-based AI agent platform, raised $10M in seed funding and reports over 100,000 agents deployed. The company positions itself as a computer-use alternative to Moltbot, the open-source agent framework that gained traction after creator Peter Steinberger's recent high-profile interview.
Steinberger built Moltbot in roughly three months working alone, using agents themselves to accelerate development. His system generated 250 billion tokens in a few months at a cost of $51,000, deployed via GitHub where users run it locally on their own machines.
The local-execution model gives Steinberger a structural advantage over incumbent AI labs. OpenAI and Anthropic cannot easily ship equivalent products because enterprise integrations depend on negotiated deals with other platforms. Moltbot, as an open-source application running on a user's own device, sidesteps those constraints. If a user logs into their New York Times account and runs Moltbot locally, the Times has limited recourse. They cannot easily block behavior tied to a user's authenticated session on their own hardware without blocking legitimate access entirely.
The licensing and liability calculus differs sharply for startups versus large labs. A corporation shipping an agent that navigates competitor platforms faces legal pressure, contract enforcement, and reputational risk. An open-source tool distributed through GitHub does not. This dynamic explains why Moltbot spread rapidly despite similar technical capabilities being available to larger players.
Twin's $10M seed and 100K+ deployed agents signal investor confidence in agentic platforms as a durable category. The company does not disclose valuation or backing details. Whether Twin's proprietary platform can compete with the open-source model that made Moltbot successful remains unclear from the available information.