AI to Reach Human-Level Intelligence Within 5-15 Years, Says OpenAI

Artificial intelligence (AI) could achieve human-level intelligence within the next five to 15 years, according to Adam D’Angelo, an OpenAI board member and CEO of Quora. D’Angelo made this prediction during a recent event, as reported by Seeking Alpha on Monday (July 29).

D’Angelo emphasized that reaching artificial general intelligence (AGI) would be a monumental shift for the world. His comments come on the heels of OpenAI’s announcement of a new five-level classification system to track progress toward AGI. Currently, OpenAI is at Level 1, where AI can engage in conversational interactions with humans. The next step, Level 2, involves AI systems capable of solving problems at the level of a human with a doctorate education.

Tracking Progress Toward AGI

Future levels envision AI systems performing extended tasks on a user's behalf, developing innovations, and eventually, at Level 5, executing the functions of an entire organization. OpenAI CEO Sam Altman and Chief Technology Officer Mira Murati have previously stated that AGI could be achieved within the next decade. Altman believes that AGI will be the most powerful tool humanity has ever created, enabling people to accomplish astonishing feats.

Implications for Commerce and Industry

Reports of OpenAI’s advancements have generated excitement in the business community about the potential for AI-powered commerce to revolutionize global trade. Ghazenfer Mansoor, founder and CEO of Technology Rivers, noted that AI’s ability to perform human-level reasoning could transform supply chain management, market trend forecasting, and customer experience personalization on a large scale.

Earlier this year, OpenAI demonstrated AI models capable of solving complex science and math questions, with one model achieving over 90% on a championship math dataset. Additionally, the company showcased a project featuring new human-like reasoning skills during an internal meeting.

Technological Innovations

Alexander De Ridder, CTO of SmythOS, explained that such AI algorithms create multiple options, follow a tree of possibilities, and reason about outcomes to select the best path, similar to how chess players strategize their moves. He suggested that OpenAI’s breakthroughs likely involve innovative algorithms that combine autonomous web research and tool usage to achieve scalable reasoning capabilities.