Augmented intelligence is the use of artificial intelligence (AI) technologies to enhance human cognitive capabilities, decision-making processes, and productivity. The label “augmented” is meant to focus attention on AI’s assistive role and emphasize the important role humans play in unlocking the complete potential of AI technologies.
The term is credited to the research firm Gartner and is often used in marketing. It is intended to shift the narrative away from the idea that AI will replace humans and focus attention on how the technology is just one more tool that humans can use to solve problems faster.
It acknowledges that while computers can do some things faster and better than humans, those skills should be used to complement human intelligence, not replace it.
|Human Intelligence||Artificial Intelligence|
|Processing Speed||Slower than machines||Faster than humans|
|Learning Ability||Capable of learning from limited data||Requires vast amounts of data to learn|
|Cognitive Flexibility||Easily adapts to new situations and can transfer learning from one context to another||Requires reprogramming and retraining to perform new tasks or the same task in different context|
|Creativity||Demonstrates high levels of creativity||Lacks innate creativity, relies on algorithms|
|Emotional Intelligence||Possesses emotional understanding and empathy||Limited ability to imitate emotions and empathy|
|Contextual Understanding||Can interpret complex and nuanced situations||Relies on predefined rules and algorithms|
|Consciousness||Experiences consciousness and self-awareness||Lacks consciousness and can only mimic self-awareness|
|Intuition||Relies on data, intuition and gut feelings||Makes predictions based on algorithmic outcomes|
Augmented Intelligence vs. Artificial Intelligence
The label augmented intelligence is sometimes considered to be a more politically correct way to describe artificial intelligence. In reality, the choice of the adjective “augmented” provides a more accurate picture of today’s AI capabilities.
That’s because the primary objective of augmented intelligence is to use AI technologies as a tool. By definition, a tool is designed for a particular task and today’s AI systems are designed to perform specific tasks.
Take ChatGPT, for example. The large language model (LLM) can generate impressive conversational responses, but it can’t create images. And while the generative AI program Dall-E excels at generating images, it can’t provide conversational responses. That’s because today’s AI is what researchers call narrow AI.
Narrow AI vs. General AI
There are two types of artificial intelligence: narrow AI and general AI.
Narrow AI (also called weak AI) refers to AI systems that are designed to perform specific tasks or functions within a limited domain. These systems excel in their designated areas but lack the versatility, creativity and adaptability of human intelligence.
In contrast, general AI (also referred to as strong AI), represents the hypothetical idea that someday, AI systems will have human-level or beyond human-level cognitive abilities. General AI is what is often depicted in science-fiction literature and films.
|Narrow AI||General AI|
|Scope||Limited to specific tasks in specific domains||Capable of performing many different types of tasks across multiple domains|
|Intelligence||Task-specific||Human-like, broad intelligence|
|Learning||Initially trained on a limited dataset||Continually trained on an unlimited variety of data sources|
|Adaptability||Limited adaptability to new scenarios||Unlimited ability to adapt to new scenarios and transfer learning to new contexts without additional training|
|Autonomy||Can perform tasks autonomously after training||Capable of generating new tasks and performing them autonomously|
At present, all AI systems, including ChatGPT, fall under the category of narrow AI. The distinguishing characteristic of narrow AI is its specificity; narrow AI systems are built and trained to excel in a designated area, and their capabilities are typically focused on addressing specific problems. General AI, with its ability to generalize and perform a wide range of tasks, remains a goal that researchers and developers are hoping to eventually achieve.