Artificial Intelligence (AI)

A branch of computer science that is focused on a machine’s capability to produce rational behavior from external inputs

Over 1.8 million professionals use CFI to learn accounting, financial analysis, modeling and more. Start with a free account to explore 20+ always-free courses and hundreds of finance templates and cheat sheets.

What is Artificial Intelligence (AI)?

Artificial Intelligence (AI) is a broad branch of computer science that is focused on a machine’s capability to produce rational behavior from external inputs. The goal of AI is to create systems that can perform tasks that would otherwise require human intelligence. AI manifests itself in everyday life via virtual assistants, search prediction technology, and even ride-hailing services.

Artificial Intelligence - Side portrait of a man and artificial intelligence concept

Key Highlights

  • Artificial Intelligence (AI) aims to develop machines that can accomplish what a human can in terms of reasoning.
  • The four types of artificial intelligence are reactive machines, limited memory, theory of mind, and self-awareness.
  • Artificial intelligence now affects productivity, employment, and competitive behavior in significant ways.

Types of Artificial Intelligence

Reactive machines

Reactive machines perceive present external information and plan actions accordingly. The machines perform specialized duties and only understand the task at hand. The machines’ behavior is consistent, given a repeated situation. In the 1990s, IBM developed a reactive machine named Deep Blue to play competitive chess, predicting chess moves by identifying each piece’s board placement.

Limited memory

Limited memory machines can harness recent observations to make informed decisions. The machines consider observational data in reference to their pre-programmed conceptual framework. The observational data is retained for a limited period and then forgotten.

Theory of mind

Theory of mind machines can form thoughts and make decisions in reference to emotional context; thus, they can participate in social interaction. The machines are still in the development stage; however, many exhibit aspects of human-like capability. For example, consider voice assistant applications that can comprehend basic speech prompts and commands but cannot hold a conversation.

Self-awareness

Self-awareness machines demonstrate intelligent behavior through ideation, the formation of desires, and understanding their internal states. In 1950, Alan Turing developed the Turing Test to identify machines that could behave indistinguishably from a human being.

Artificial Intelligence - Types of AI

Artificial Intelligence’s Increasing Relevance

Artificial intelligence’s growing popularity in the 21st century is largely due to the advancements in the sub-field of machine learning. Machine learning develops systems that improve upon themselves, which is accomplished through the identification of algorithms. Some processes that machine learning optimizes include paperwork automation, forensic accounting, and algorithmic trading.

How is Machine Learning Achieved?

The easiest way to understand how a machine becomes intelligent with AI is to compare it to how humans learn. For example, consider a child learning how to ride a bicycle. The child mounts the bicycle, grips the handlebars, and hopes to stay upright and in control. The child does not learn how to ride a bike by understanding the physics of biking but rather through trial and error.

Over time, the child becomes instinctively adept in perceiving factors that can make him lose control of the bicycle. Just as a child learns the unwritten rules of riding a bicycle through practice, artificial intelligence is developed through repeated simulation.

Supervised learning

Supervised learning is the most common learning method in the field of artificial intelligence. A machine attempts to derive a function given labeled sets of input and output pairs. When dealing with a numerical data set, regression is used. When dealing with categorical variables, classification is the preferred method. If the model provides an incorrect answer, the model can be adjusted to provide more accurate outputs.

Unsupervised learning

Unsupervised learning involves a machine transforming data into useful information. Common methods include clustering and association. Clustering groups similar variables together, whereas association detects correlation among variables. Data mining utilizes clustering and association to filter through large data sets. The process of transforming large data sets into meaningful information can be optimized with unsupervised learning.

Reinforcement learning

In reinforcement learning, a machine autonomously responds to external stimuli and is conditioned through occasional rewards and punishment. The purpose of the reinforcement learning method is to develop a machine that can act rationally independently.

The large investment bank, J.P. Morgan, currently uses reinforcement learning algorithms to place trades. This is accomplished through programming that accordingly awards or penalizes the algorithm depending on the decision made.

Future Impact of Artificial Intelligence

Rapid advances in artificial intelligence will result in a profound impact on productivity, employment, and competition. However, AI’s future integration into society is a controversial subject.

Impact on productivity

Increasing economic productivity leads to more satisfied customers and strengthened corporate profitability. In the airline industry, AI will drive customer satisfaction through accurately scheduled and safer flights. Businesses that harness AI can improve their value proposition to customers while improving profitability at the same time.

For example, Delta Airlines leverages machine learning to provide its customers with a superior flying experience. Delta analyzes Big Data to learn about aircraft positions, weather conditions, and aircraft diagnostics. Hypothetical outcomes and their probabilities are then identified. The airline then optimizes its flight scheduling regarding the potential outcomes.

Impact on employment

Rapid expansion in the artificial intelligence field will result in more high-paying jobs, which, in turn, will require more highly-educated employees. The largest criticism of AI is that it will automate low-skill jobs and increase the unemployment rate for less-educated people.

Furthermore, low-skilled employees are more likely to be minorities due to systematic discrimination. It is, therefore, argued that artificial intelligence could reinforce systematic discriminatory practices.

Impact on competition

Companies attempting to achieve a competitive advantage can leverage AI to optimize their business. Currently, only innovators and a few early adopters are integrating AI into their businesses. Once the economic benefit of AI integration outweighs the R&D and integration costs, more companies will adopt the new technology.

Assume a toy manufacturer decides to use machine learning to optimize its supply chain further. The manufacturer can take advantage of its lowered cost of goods sold and offer toys at a lower price point to capture market share. This demonstrates how AI can increase competitive behavior in the marketplace.

Additional Resources

Artificial Intelligence in Finance and Banking

Will ChatGPT Be My Banker?

How Finance and Banking Professionals Can Use ChatGPT

See all data science resources

0 search results for ‘