AI Rebrand: Trump Pushes 'Super Intelligence'

In recent weeks, U.S. President Donald Trump has initiated a significant debate regarding artificial intelligence, advocating for its rebranding as 'super intelligence.' This rebranding effort, formalized through an executive order on September 29 and the establishment of the Super Intelligence Force for AI policy, aims to underscore the rapidly advancing capabilities of these technologies and cement U.S. leadership.

The push echoes sentiments from figures like Meta CEO Mark Zuckerberg and philosopher Nick Bostrom, who have also discussed 'superintelligence.' However, the article distinguishes Trump's concept, focused on current capabilities and national dominance, from Bostrom's vision of an intellect far surpassing human cognitive abilities across all domains.

Historically, AI discourse distinguished between 'weak AI' (simulations) and 'strong AI' (actual cognition), later evolving into 'narrow AI' (task-specific) and 'artificial general intelligence' (AGI) (human-level performance across tasks). AGI, a goal for companies like DeepMind and OpenAI, is now a significant business model, though its definition remains somewhat vague.

Despite AGI's pursuit, the article contends that true 'superintelligence' remains distant. Challenges persist in defining and measuring intelligence itself, with traditional IQ tests proving problematic for AI. Newer AI-specific tests like ARC-AGI offer alternative metrics, but definitions and evaluation standards are in constant flux.

The article concludes that predicting the arrival of AGI or superintelligence is difficult. The rapid evolution of AI means that what seems like a breakthrough today may become commonplace or even flawed tomorrow, suggesting that if AGI arrives, it might be a gradual realization rather than a singular event, potentially under a new name.