At the Financial Times Future of AI Summit in London, a group of prominent artificial intelligence researchers and industry leaders declared that AI systems have reached what they consider to be human-level intelligence in certain domains.

The announcement, made by figures including Nvidia CEO Jensen Huang, Meta’s Chief AI Scientist Yann LeCun, Yoshua Bengio, Geoffrey Hinton, Fei-Fei Li, and Bill Dally, reignited global debate about whether artificial general intelligence (AGI) has already arrived.

Huang said, 'For the first time, AI is intelligence that augments people, it addresses labour, it does work.

We have enough general intelligence to translate the technology into an enormous amount of society-useful applications in the coming years; we are doing it today.' His remarks were echoed by LeCun, who emphasized that AGI will not be a sudden event but a gradual expansion of capabilities across domains.

Bengio, while cautious, added that 'we are already there'—suggesting that human-level performance in some tasks may no longer be hypothetical.

The claim comes as AI systems continue to surpass human benchmarks in translation, pattern recognition, and data reasoning.

However, experts caution that these achievements remain limited to specific, well-defined areas.

True AGI—defined as the ability to understand, learn, and apply knowledge across all contexts, including emotional and moral reasoning—remains a point of contention.

A recent AGI Progress Report found that while modern systems outperform humans in narrow tasks, they still lack the breadth and adaptability of full human cognition.

Analysts note that what qualifies as 'human-level' is itself ambiguous: does it mean matching average human ability in individual tests, or achieving the flexible, context-aware intelligence that defines human thought?

The broader implications of this moment are profound.

Investment in AI companies has surged, with references to AGI in corporate filings and earnings calls increasing more than fifty percent year-on-year in early 2025.

Governments are now revisiting definitions of intelligence and rethinking regulatory frameworks that may no longer capture the pace or scope of machine learning advances.

The consensus among experts at the summit was clear: whether or not true AGI has arrived, the world must prepare for systems that are already powerful enough to reshape industries, economies, and societies.

The shift in tone—from 'someday' to 'already'—marks a turning point in how humanity perceives its technological reflection.