Several pioneers in the AI field have stated that AI systems have already achieved intelligence comparable to humans in certain domains, intensifying debates about the timeline for "superintelligence." These experts argue that artificial general intelligence (AGI) will emerge gradually rather than at a single defining moment.

NVIDIA CEO Jensen Huang, Meta AI chief Yann LeCun, and other leading computer scientists were jointly awarded this year's Queen Elizabeth Prize for Engineering. During their acceptance speeches at the Financial Times AI Future Summit, they noted that machines have attained "human-level intelligence" in specific areas. Huang remarked, "For the first time, AI has become intelligence that augments human capability—it solves labor challenges and completes tasks."

These statements have prompted a reassessment of valuation metrics for AI companies. Firms like OpenAI and Anthropic have attracted billions in investments, partly based on expectations of their potential to develop world-changing technologies. Data shows mentions of AGI in Q1 2025 earnings calls increased 53% year-over-year.

Market expectations for AGI's arrival remain sharply divided. The most optimistic researchers and investors predict its emergence within two years, while others believe this milestone remains decades away. These diverging views directly impact investment trends and company valuations in the AI sector.

Is AGI Already Here or Still Distant?

The award-winning experts maintain that AGI won't arrive as a sudden breakthrough. LeCun stated, "Capabilities will expand progressively across domains—this won't be a single event." Huang was more direct: "We're already there... It's becoming somewhat academic because the technology will continue being applied." He added that current general intelligence levels already enable transformative societal applications.

Achieving AGI has become one of the most pressing challenges in the rapidly evolving AI industry. The race between OpenAI and Anthropic has drawn billions in funding, with both the U.S. and China competing to reach this goal first. Recent valuation surges for public and private AI companies partly reflect belief in their potential to develop revolutionary technologies.

Will Machines Surpass Humans Completely?

Experts disagree about whether AI systems will eventually outperform humans across all domains. World Labs founder Fei-Fei Li observed, "Some machines will exceed human intelligence... portions are already here. How many of us can recognize 22,000 objects... or translate 100 languages?" However, she emphasized that "machine-based intelligence will accomplish powerful things, but human intelligence will always hold crucial societal importance."

Geoffrey Hinton, last year's Nobel Physics laureate, predicted: "When debating machines, they'll always win—I'm certain this will happen within 20 years." Turing Award winner Yoshua Bengio added, "Eventually we'll build machines capable of nearly everything humans do—I see no conceptual barrier." However, Bengio cautioned against present decisions based on speculative futures: "Maintain agnosticism—avoid grand declarations when multiple outcomes remain possible." This measured perspective contrasts with current market enthusiasm about AGI, offering investors room for sober reflection.