A recent report from The Information has shed light on the internal definition of artificial general intelligence (AGI) used by Microsoft and OpenAI, and it's not what many in the tech community expected. According to the report, AGI is defined by the startup's profits, specifically when OpenAI develops AI systems that can generate at least $100 billion in profits.

This definition deviates significantly from the rigorous technical and philosophical understanding of AGI, which is often characterized by a machine's ability to perform any intellectual task that a human can. Instead, Microsoft and OpenAI's definition is centered around financial returns, raising questions about the motivations behind their partnership and the direction of AI development.

The report comes at a time when OpenAI is reportedly set to lose billions of dollars this year, with the startup telling investors that it won't turn a profit until 2029. This timeline is crucial, as Microsoft loses access to OpenAI's technology when the startup reaches AGI, as defined by their agreement. This means that Microsoft could have access to OpenAI's models for a decade or more, giving the tech giant a significant advantage in the AI race.

The implications of this definition are far-reaching, and some have speculated that OpenAI might declare AGI sooner rather than later to limit Microsoft's access to its technology. However, this agreement suggests that OpenAI is unlikely to do so, as it would mean giving up control over its models and potentially sacrificing its long-term goals.

Last week, the AI community debated whether OpenAI's o3 model was a meaningful step toward AGI. While o3 may perform better than other AI models, it also comes with significant compute costs, which could be a major hurdle for OpenAI and Microsoft's profit-centric definition of AGI. The high costs associated with developing and deploying AI models could make it challenging for OpenAI to reach the $100 billion profit threshold, potentially delaying its achievement of AGI.

The partnership between Microsoft and OpenAI has been instrumental in driving AI innovation, but this new definition of AGI raises questions about the motivations behind their collaboration. Is the primary goal of AI development to generate profits, or is it to create machines that can truly think and act like humans? As the AI landscape continues to evolve, it's essential to re-examine the values and priorities that drive this technology forward.

In conclusion, the revelation of Microsoft and OpenAI's internal definition of AGI highlights the complex and often conflicting interests that shape the development of AI. As the tech community continues to grapple with the implications of this definition, it's clear that the path to true AGI is fraught with challenges, uncertainties, and competing priorities.