The proliferation of agentic artificial intelligence in digital markets presents an unprecedented challenge to established competition law frameworks. Unlike conventional pricing software, agentic AI systems powered by deep reinforcement learning and Q-learning autonomously observe market conditions, adapt strategies, and converge on supra-competitive equilibria without any explicit human instruction or inter-firm communication. This paper argues that this technological evolution creates a critical enforcement vacuum: foundational antitrust statutes, including Section 1 of the Sherman Act (US) and Article 101 TFEU (EU), are premised on identifying a human ‘meeting of the minds’ or explicit agreement, evidentiary standards that are wholly inadequate when collusion emerges as a machine-discovered, profit-maximising strategy. Drawing on Ezrachi and Stucke’s taxonomy of algorithmic collusion messenger, hub-and-spoke, predictable agent, and digital eye the paper demonstrates that existing jurisprudence across the United States, European Union, United Kingdom, and India successfully captures only the first two categories, while remaining structurally blind to emergent, autonomous collusion. Through an interdisciplinary analytical framework combining antitrust law, game theory, and computer science, the paper further identifies three core doctrinal failures: the impossibility of proving explicit intent against a black-box algorithm, an unresolved agency and liability gap between developers and deployers, and the risk of ‘superhuman collusion’ that surpasses the durability of any human cartel. To remedy these failures, this paper proposes a two-pronged normative framework: first, the formal recognition of ‘algorithmic agreement’ as a distinct category of anti-competitive conduct, shifting the evidentiary burden from intent to sustained, machine-driven parallel pricing outcomes; and second, a hybrid liability model holding both developers and deployers accountable across the algorithmic supply chain. These proposals are complemented by recommendations for mandatory ex-ante algorithmic auditing, regulatory sandboxes, and international harmonisation through the OECD and ICN, alongside specific legislative amendments to India’s Competition Act, 2002.