This chapter addresses a core theoretical paradox in sustainable entrepreneurship, first identified by Blok (Sustainable business models: Principles, promise, and practice, Springer, 2018): how can entrepreneurs reduce information asymmetries—essential for ecosystem collaboration—while preserving those that underpin competitive advantage? Drawing on organizational paradox theory, we develop two constructs that reframe information dynamics in AI-mediated settings. First, Algorithm-Mediated Collaboration (AMC) leverages the computational malleability of information to sustain transparency and confidentiality simultaneously. Second, Computational Epistemic Sufficiency (CES) redefines the conditions for collective action, showing how AI-augmented collective intelligence can meet knowledge thresholds without full transparency. Our framework specifies six transformative mechanisms that reconfigure competition–cooperation relations, including the decoupling of engagement from vulnerability and the resolution of the innovation–appropriation paradox. These contributions extend organizational paradox theory by positioning AI as a paradoxical mediator that maintains productive tensions while enabling action. Grounded in organizational paradox theory and design principles for AI-mediated collaboration, we formulate testable propositions and provide operational guidance (indicators, data sources, methods) to support evidence-based uptake. We also outline practical implications for entrepreneurs (informational orchestration capabilities), policymakers (responsive regulation), and support organizations (algorithmic literacy). While resolving traditional informational tensions, AI also introduces new paradoxes, requiring reflexive governance.