Jensen Huang, CEO of Nvidia, stated that the launch of OpenAI's GPT-6 Astra represents the arrival of Artificial General Intelligence (AGI), which is considered the primary goal of the largest companies in the AI sector.
In a post on the X platform, the executive congratulated the OpenAI team on the results achieved and emphasized the progression of models over the last four years, categorically stating: 'AGI has arrived.' Huang also highlighted Nvidia's own contribution to this development.
AGI, generically defined as a system capable of performing various intellectual tasks with performance equal to or superior to humans, lacks a universally accepted test for its determination, leading to different interpretations of its scope.
For OpenAI itself, AGI consists of a highly autonomous system capable of surpassing humans in most economically valuable activities. The company believes that GPT-6 Astra, launched last week, initiated this new technological phase.
According to Business Insider reports, Greg Brockman commented that this new model may be seen in the future as a milestone in the emergence of AGI, as it introduces a strong focus on task execution as an agent, enhancing autonomy in tool usage.
However, the absence of defined criteria for this technological advance raises questions about the exact moment such a leap will occur. Researcher Gary Marcus criticized Huang for not providing concrete proof to validate the Astra launch as a milestone, as claimed by OpenAI.
Marcus argued that, using standards established in computer science, Astra satisfies only one or two requirements, indicating that the model is still far from the level required to be classified as AGI.
Sam Altman, CEO of OpenAI, expressed the belief that the term AGI has an imprecise definition and primarily functions as a marketing concept. Although he reinforced the arrival of AGI, Brockman also observed that the term lost some of its importance after the end of contractual clauses granting Microsoft exclusivity over OpenAI's technologies until AGI was reached.
This is not the first time Jensen Huang has addressed the topic; in March, he already argued that AIs would be capable of managing, scaling, and creating a technology company valued at over US$1 billion (equivalent to approximately R$5.1 billion), considering this a sign that AI had reached a higher level.
However, Marcus interpreted Huang's optimism as an 'attempt to improperly appropriate a scientific issue by corporate decree.'
Large Language Model (LLM) developing companies, such as OpenAI, rely on vast volumes of accelerators to train and operate their own models, with Nvidia being a crucial supplier of this hardware.
The high demand for computational capacity is also reflected in Nvidia's financial results. In the balance sheet released in August, the company reported a quarterly revenue of US$96.2 billion (about R$493 billion), more than double the amount recorded in the same period last year. Of this amount, US$89 billion (R$456 billion) came from the data center and AI chip segment.
In the same disclosure where he confirmed the arrival of AGI, Huang announced that another 400,000 Nvidia graphics processors would come online in the global network.
