At the company’s annual developer conference, Jensen Huang asserted that Nvidia’s latest AI models now match human cognitive tasks in accuracy and adaptability, marking what he described as the arrival of human-level AI.

Despite the breakthrough claim, Huang highlighted ongoing challenges in model robustness, real-world generalization and energy efficiency, indicating that wide-scale commercial deployment will require further R&D and system validation.

Following the remarks, analysts adjusted their near-term expectations for AI-driven revenue growth, balancing optimism over potential new use cases with caution about the pace of commercial adoption and capital expenditure needs.