Why comparisons between AI and human intelligence miss the point

2026-02-05 · Source: Artificial intelligence (AI) – The Conversation · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Emerging Technologies & Innovation · Depth: Intermediate, short

Summary

Claims of imminent AI superintelligence, often based on large language model advancements, frequently compare AI to individual human cognitive performance. However, this perspective overlooks the fundamentally social, embodied, and collective nature of human intelligence. Research in cognitive science and anthropology demonstrates that human achievements, including scientific discovery and language, emerge from collective processes like shared language, cultural transmission, and cooperation across generations. AI systems, conversely, process information in isolation, lacking the capacity for cooperation, social bonding, or shared moral reasoning. Furthermore, AI's reliance on vast but biased datasets, predominantly in a few languages, limits its access to the diverse cognitive landscape of humanity. The article also highlights data scarcity as a looming constraint, with training on AI-generated data risking amplified errors and biases, creating an "echo chamber" rather than deeper understanding.

Key takeaway

For AI researchers and ethicists evaluating claims of superintelligence, you should critically assess the underlying definition of intelligence. Recognize that comparing AI to isolated human cognitive performance is a "category error" that ignores the collective, embodied, and social foundations of human intellect. Focus on integrating AI as a useful tool with careful oversight, addressing real issues like bias and governance, rather than being swayed by exaggerated expectations of machine supremacy.

Key insights

Human intelligence is fundamentally collective, embodied, and social, a dimension AI systems currently lack.

Principles

• Collective intelligence often outperforms individual brilliance.
• Embodied experience grounds human abstract reasoning.
• Data diversity is crucial for robust intelligence.

In practice

• Evaluate AI against collective human capabilities.
• Prioritize diverse, high-quality human data for training.
• Recognize AI as a tool, not a competing mind.

Topics

• AI Superintelligence
• Collective Intelligence
• Embodied Cognition
• AI Data Bias
• Data Scarcity

Best for: AI Researcher, AI Ethicist, Tech Journalist

Related on AIssential

• AI is intelligent. Humans are informed. That difference still matters. — Humans excel over AI in tasks requiring nuanced contextual information inaccessible to current models.
• Import AI 451: Political superintelligence; Google's society of minds, and a robot drummer — AI's societal impact hinges on intentional design, governance, and integration within complex human-AI ecosystems.
• The Shape of Artificial Intelligence — AI and human intelligence are both "spiky stars" of capabilities, not smooth circles, due to distinct optimization processes.
• The intelligence illusion: why AI isn’t as smart as it is made out to be — AI's perceived intelligence is an illusion stemming from anthropomorphization and conflating information processing with human cognition.
• "How to measure intelligence?" | Six researchers debate — Current AI lacks human-like developmental learning, struggling with common sense despite advanced problem-solving.
• Artificial general intelligence must be assessed in its scientific and societal context — Assessing AGI requires considering both scientific and societal contexts, not just technical capabilities.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Artificial intelligence (AI) – The Conversation.