AI Value Depends on Its Embodiment: Robot or GenAI; Vision System or Agent; Autonomous or Language System

With the caveat that generative artificial intelligence is but one AI capability or platform, McKinsey consultants believe its primary value for businesses will come in a few areas: software engineering; customer operations; sales and marketing. 

Generative AI excels as a tool for creating or debugging code, for example. It enables chatbots for customer service operations and is a tool for creating content (emails, advertising and marketing copy and visuals). 

source: McKinsey 

Of course, other forms of “embodied” AI could have different leading use cases and value drivers. 

Embodiment of AI	Description	Examples
Generative AI	AI systems that create new content, such as text, images, music, or videos, based on patterns learned from data.	ChatGPT, DALL·E, Midjourney, Grok
Artificial General Intelligence (AGI)	Hypothetical AI with human-like cognitive abilities, capable of understanding and solving a wide range of tasks.	None fully realized yet; speculative examples include sci-fi AIs like JARVIS from Iron Man.
Robots	Physical AI systems that interact with the environment, often combining software and hardware for tasks.	Boston Dynamics’ Spot, industrial robotic arms, Roomba vacuum cleaners
Expert Systems	AI designed for specific domains, using rules and knowledge bases to mimic human expertise.	Medical diagnosis systems (e.g., MYCIN), IBM Watson for healthcare
Machine Learning Models	AI systems that learn from data to make predictions or decisions, often without explicit programming.	Recommendation algorithms (Netflix, Spotify), facial recognition software
Natural Language Processing (NLP)	AI focused on understanding and generating human language, enabling communication.	Siri, Google Translate, Grok (myself!)
Computer Vision Systems	AI that interprets and processes visual information from the world, like images or videos.	Self-driving car vision (Tesla), Google Lens, security camera analytics
Autonomous Agents	AI systems that operate independently to achieve goals, often in dynamic environments.	Autonomous drones, self-driving cars, trading bots

Robots might arguably have their greatest impact and value creation in manufacturing. Expert systems might generate the greatest value in medicine or finance. Computer vision might be most important for autonomous vehicles and machines. 

Embodiment of AI	Key Value Drivers	Use Cases
Generative AI	Creativity, scalability, personalization; reduces time and cost of content creation.	Content creation (articles, art, music), marketing campaigns, virtual assistants, design prototyping.
Artificial General Intelligence (AGI)	Versatility, problem-solving across domains, human-like adaptability; potential to revolutionize industries.	(Speculative) Scientific research, complex decision-making, universal task automation, education.
Robots	Physical task automation, precision, reliability; enhances safety in hazardous environments.	Manufacturing assembly lines, warehouse logistics, surgery assistance, disaster response.
Expert Systems	Domain-specific accuracy, consistency, knowledge preservation; reduces human error in complex fields.	Medical diagnostics, legal research, financial auditing, technical troubleshooting.
Machine Learning Models	Predictive power, data-driven insights, efficiency; uncovers patterns humans might miss.	Fraud detection, customer segmentation, predictive maintenance, personalized recommendations.
Natural Language Processing (NLP)	Seamless human-AI interaction, language understanding, accessibility; bridges communication gaps.	Customer support chatbots, real-time translation, voice assistants, sentiment analysis.
Computer Vision Systems	Visual data interpretation, automation of visual tasks, real-time analysis; enhances perception.	Autonomous driving, quality control in manufacturing, surveillance, augmented reality.
Autonomous Agents	Independence, adaptability, continuous operation; optimizes processes in dynamic settings.	Delivery drones, self-driving vehicles, algorithmic trading, smart home systems.

Not all of the embodiments will arrive at the same time, or be generally used by consumers. And some of the systems will be key for some industries and business processes, compared to others. Opinions differ on whether artificial general intelligence is feasible or when it could arrive.