OpenAI's Vision for Artificial Superintelligence

OpenAI’s CEO, Sam Altman, has recently expressed confidence in the company’s ability to build artificial general intelligence (AGI). He mentioned that OpenAI knows "how to build AGI as we have traditionally understood it," a term indicating the level of intelligence that matches or exceeds human capabilities.

In a blog post released on a Monday, Altman discussed OpenAI's journey and its aspirations for the future. The next significant goal for the organization is achieving "superintelligence in the true sense of the word." According to him, superintelligent AI tools have the potential to significantly enhance scientific exploration and innovation, far beyond human capabilities. He believes that this progress could lead to increased abundance and prosperity for society.

Altman’s understanding of superintelligence seems to align closely with OpenAI's prior definition of AGI, which is described as AI that is generally smarter than humans. OpenAI has consistently emphasized its foundational mission to create AGI that benefits humanity. However, Altman recently attempted to temper expectations surrounding AGI, suggesting that it may not have the impact many envision.

This adjustment in expectations might be advantageous for OpenAI, especially given its financial relationship with Microsoft. The company has exclusivity agreements with Microsoft, which means that OpenAI cannot officially claim it has achieved AGI until certain benchmarks are met. Nevertheless, reports indicate that Microsoft views AGI as a system capable of generating substantial profits, complicating the situation for OpenAI.

Though there have been rumors that OpenAI plans to combine its large language models into a unified "AGI" product to reach its goals more swiftly, the company is currently not operating profitably. Altman mentioned that OpenAI is facing losses, particularly with subscriptions like the ChatGPT Pro plan, which costs $200 per month.

In a series of posts on X, Altman shared, "People use it much more than we expected. I personally chose the price and thought we would make some money." However, he did not specifically address the complicated profit-sharing arrangement between OpenAI and Microsoft in his blog.

Altman also reflected on the events that led to his departure from OpenAI as CEO, his brief stint with Microsoft, and his subsequent return in November 2023. Since his return, he has consolidated leadership at OpenAI, which is shifting from a non-profit to a for-profit model this year.

He characterized the earlier governance issues as a "failure by well-meaning people," including himself. Altman expressed regret and a desire for improved leadership practices. He stated, "Good governance requires a lot of trust and credibility. I appreciate the progress we’ve made in establishing a stronger governance system at OpenAI that can support our mission to ensure AGI benefits all of humanity."

OpenAI, Altman, superintelligence