Decoding the Neural Code: How AI Will Overtake Human Intelligence
In “Towards Human-Level Artificial Intelligence,” Eitan Michael Azoff suggests that deciphering the neural code and replicating visual thinking are essential steps for AI to surpass human intelligence.
According to an AI expert, grasping how ‘visual thinking’ functions is crucial for developing human-level AI. He suggests that computers might be capable of mimicking a form of consciousness but also cautions that society must regulate AI technology and maintain exclusive control over the ‘off switch.’
AI Advancement Through Neural Code Decoding
According to AI technology analyst Eitan Michael Azoff, humans will create AI that surpasses our abilities once we decipher the ‘neural code.’ Azoff explains that this breakthrough will enable us to engineer intelligence with greater speed and capacity than the human brain.
The key to this advancement lies in understanding how the brain encodes sensory information and processes it for cognitive functions like thinking, learning, problem-solving, internal visualization, and internal dialogue.
In his book Towards Human-Level Artificial Intelligence: How Neuroscience can Inform the Pursuit of Artificial General Intelligence, Eitan Michael Azoff argues that a key step in creating ‘human-level AI’ is replicating consciousness in computers.
AI Advancement Through Neural Code Decoding:
Azoff points out that there are various forms of consciousness, with scientists recognizing that even simpler creatures like bees possess some level of consciousness, though without self-awareness. Humans experience something similar when fully immersed in a task, often referred to as being “in the flow.”
He suggests that a virtual brain could be created through computer simulation, which could initially mimic consciousness without self-awareness. This type of consciousness allows animals to plan actions, anticipate events, and recall past experiences, and it could serve the same purpose for AI.
Visual thinking may hold the key to understanding the nature of consciousness. While current AI relies on large language models (LLMs) and doesn’t “think” visually, Eitan Michael Azoff notes that visual thinking came before language in human evolution. He suggests that unraveling visual thinking and modeling visual processing will be a vital step in developing human-level AI.
Azoff explains, Once we crack the neural code, we will create faster and more advanced brains with greater capacity, speed, and enhanced technology that will surpass the human brain.
“We will begin by modeling visual processing, which will allow us to replicate visual thinking. I believe that from this, a form of ‘in-the-flow’ consciousness will emerge. I do not think a system needs to be alive to possess consciousness.”
Azoff also offers a caution, urging society to regulate AI technology to prevent potential misuse. He emphasizes, Until we can fully trust the machines we create, two key measures must always be in place.
“First, humans must retain sole control over the off switch. Second, AI systems should be built with embedded behavior safety rules.”
Read the original article on: ScitechDaily
Read more: AI Generates More Innovative Research Ideas than Human Experts
