As soon as OpenAI's Astra model nearly 'solved' the ARC AGI 3 benchmark, its creator immediately 'moved the goalposts' by announcing the next version will focus on 'open-ended invention.' This shows how the very definition of AGI is a moving target, constantly being redefined by technological breakthroughs.

Mustafa Suleiman offers clear definitions: AGI is human parity on most tasks. Superintelligence exceeds human performance and discovers new knowledge. The Singularity is the sci-fi point where a superintelligence can recursively self-improve. This clarifies the ladder of AI progression beyond generic terms.

OpenAI's CEO believes the term "AGI" is ill-defined and its milestone may have passed without fanfare. He proposes focusing on "superintelligence" instead, defining it as an AI that can outperform the best human at complex roles like CEO or president, creating a clearer, more impactful threshold.

"Superintelligence" is clearly defined as AI that is better, faster, and cheaper than the best humans at everything. In contrast, "AGI" (Artificial General Intelligence) is a vague term for general-purpose AI, a milestone that current models have arguably already achieved.

OpenAI's previous structure allowed it to potentially end Microsoft's IP rights by declaring AGI. By removing this clause in the new deal, OpenAI has given up its most powerful negotiating lever, signaling that AGI is not a near-term reality or that the partnership was more critical than the threat.

Microsoft’s new superintelligence team is a direct result of a renegotiated OpenAI deal. The previous contract restricted Microsoft from building AGI past a certain computational threshold. Removing this clause was a pivotal, strategic move to pursue AI self-sufficiency.

The ultimate goal for companies like OpenAI and Anthropic is not just creating useful products like chatbots, but developing superintelligence—an AI that surpasses human cognitive ability in every domain, akin to the gap between a human and a mouse.

OpenAI and Microsoft operate with a clear, economically-focused definition of AGI: a system outperforming humans at most economically valuable work. In contrast, Google's leadership admits they are still 'working on' a definition. This ambiguity reveals a lack of unified strategic direction and focus compared to its rivals.

AI companies exploit the lack of a scientific consensus on 'AGI' (Artificial General Intelligence) by defining it differently to suit their audience—as a cure-all for regulators, a helpful assistant for consumers, or a revenue machine for investors.