Artificial intelligence – How to Get Help

Artificial intelligence (AI) is a branch of computer science that aims to build machines capable of performing tasks that typically require human intelligence. AI enables machines to simulate human abilities, such as learning, problem-solving, decision-making and comprehension. Common applications of AI include speech recognition, image recognition, content generation, recommendation systems and self-driving cars. While AI is an interdisciplinary science with multiple approaches, advancements in machine learning and deep learning in particular are changing virtually every industry, making AI an increasingly integral part of everyday life.

What Is Artificial Intelligence?

Artificial intelligence refers to computer systems that are capable of performing tasks traditionally associated with human intelligence — such as making predictions, identifying objects, interpreting speech and generating natural language. AI systems learn how to do so by processing massive amounts of data and looking for patterns to model in their own decision-making. In many cases, humans will supervise an AI’s learning process, reinforcing good decisions and discouraging bad ones, but some AI systems are designed to learn without supervision.

Over time, AI systems improve on their performance of specific tasks, allowing them to adapt to new inputs and make decisions without being explicitly programmed to do so. In essence, artificial intelligence is about teaching machines to think and learn like humans, with the goal of automating work and solving problems more efficiently.

Artificial intelligence is the simulation of human intelligence processes by machines, especially computer systems. Examples of AI applications include expert systems, natural language processing (NLP), speech recognition and machine vision.

As the hype around AI has accelerated, vendors have scrambled to promote how their products and services incorporate it. Often, what they refer to as “AI” is a well-established technology such as machine learning.

AI requires specialized hardware and software for writing and training machine learning algorithms. No single programming language is used exclusively in AI, but Python, R, Java, C++ and Julia are all popular languages among AI developers.

All but the simplest human behavior is ascribed to intelligence, while even the most complicated insect behavior is usually not taken as an indication of intelligence. What is the difference? Consider the behavior of the digger wasp. When the female wasp returns to her burrow with food, she first deposits it on the threshold, checks for intruders inside her burrow, and only then, if the coast is clear, carries her food inside. The real nature of the wasp’s instinctual behavior is revealed if the food is moved a few inches away from the entrance to her burrow while she is inside: on emerging, she will repeat the whole procedure as often as the food is displaced. Intelligence—conspicuously absent in the case of the wasp—must include the ability to adapt to new circumstances.

How Does AI Work?

Artificial intelligence systems work by using algorithms and data. First, a massive amount of data is collected and applied to mathematical models, or algorithms, which use the information to recognize patterns and make predictions in a process known as training. Once algorithms have been trained, they are deployed within various applications, where they continuously learn from and adapt to new data. This allows AI systems to perform complex tasks like image recognition, language processing and data analysis with greater accuracy and efficiency over time.

AI systems work by ingesting large amounts of labeled training data, analyzing that data for correlations and patterns, and using these patterns to make predictions about future states.

For example, an AI chatbot that is fed examples of text can learn to generate lifelike exchanges with people, and an image recognition tool can learn to identify and describe objects in images by reviewing millions of examples. Generative AI techniques, which have advanced rapidly over the past few years, can create realistic text, images, music and other media.

Programming AI systems focuses on cognitive skills such as the following:

Learning. This aspect of AI programming involves acquiring data and creating rules, known as algorithms, to transform it into actionable information. These algorithms provide computing devices with step-by-step instructions for completing specific tasks.
Reasoning. This aspect involves choosing the right algorithm to reach a desired outcome.
Self-correction. This aspect involves algorithms continuously learning and tuning themselves to provide the most accurate results possible.
Creativity. This aspect uses neural networks, rule-based systems, statistical methods and other AI techniques to generate new images, text, music, ideas and so on.

Machine Learning

The primary approach to building AI systems is through machine learning (ML), where computers learn from large datasets by identifying patterns and relationships within the data. A machine learning algorithm uses statistical techniques to help it “learn” how to get progressively better at a task, without necessarily having been programmed for that certain task. It uses historical data as input to predict new output values. Machine learning consists of both supervised learning (where the expected output for the input is known thanks to labeled data sets) and unsupervised learning (where the expected outputs are unknown due to the use of unlabeled data sets).

Neural Networks

Machine learning is typically done using neural networks, a series of algorithms that process data by mimicking the structure of the human brain. These networks consist of layers of interconnected nodes, or “neurons,” that process information and pass it between each other. By adjusting the strength of connections between these neurons, the network can learn to recognize complex patterns within data, make predictions based on new inputs and even learn from mistakes. This makes neural networks useful for recognizing images, understanding human speech and translating words between languages.

Deep Learning

Deep learning is an important subset of machine learning. It uses a type of artificial neural network known as deep neural networks, which contain a number of hidden layers through which data is processed, allowing a machine to go “deep” in its learning and recognize increasingly complex patterns, making connections and weighting input for the best results. Deep learning is particularly effective at tasks like image and speech recognition and natural language processing, making it a crucial component in the development and advancement of AI systems.

Natural Language Processing

Natural language processing (NLP) involves teaching computers to understand and produce written and spoken language in a similar manner as humans. NLP combines computer science, linguistics, machine learning and deep learning concepts to help computers analyze unstructured text or voice data and extract relevant information from it. NLP mainly tackles speech recognition and natural language generation, and it’s leveraged for use cases like spam detection and virtual assistants.

Computer Vision

Computer vision is another prevalent application of machine learning techniques, where machines process raw images, videos and visual media, and extract useful insights from them. Deep learning and convolutional neural networks are used to break down images into pixels and tag them accordingly, which helps computers discern the difference between visual shapes and patterns. Computer vision is used for image recognition, image classification and object detection, and completes tasks like facial recognition and detection in self-driving cars and robots.

Why Is Artificial Intelligence Important?

Artificial intelligence aims to provide machines with similar processing and analysis capabilities as humans, making AI a useful counterpart to people in everyday life. AI is able to interpret and sort data at scale, solve complicated problems and automate various tasks simultaneously, which can save time and fill in operational gaps missed by humans.

AI serves as the foundation for computer learning and is used in almost every industry — from healthcare and finance to manufacturing and education — helping to make data-driven decisions and carry out repetitive or computationally intensive tasks.

Many existing technologies use artificial intelligence to enhance capabilities. We see it in smartphones with AI assistants, e-commerce platforms with recommendation systems and vehicles with autonomous driving abilities. AI also helps protect people by piloting fraud detection systems online and robots for dangerous jobs, as well as leading research in healthcare and climate initiatives.

AI is important for its potential to change how we live, work and play. It has been effectively used in business to automate tasks traditionally done by humans, including customer service, lead generation, fraud detection and quality control.

In a number of areas, AI can perform tasks more efficiently and accurately than humans. It is especially useful for repetitive, detail-oriented tasks such as analyzing large numbers of legal documents to ensure relevant fields are properly filled in. AI’s ability to process massive data sets gives enterprises insights into their operations they might not otherwise have noticed. The rapidly expanding array of generative AI tools is also becoming important in fields ranging from education to marketing to product design.

Advances in AI techniques have not only helped fuel an explosion in efficiency but also opened the door to entirely new business opportunities for some larger enterprises. Prior to the current wave of AI, for example, it would have been hard to imagine using computer software to connect riders to taxis on demand, yet Uber has become a Fortune 500 company by doing just that.

AI has become central to many of today’s largest and most successful companies, including Alphabet, Apple, Microsoft and Meta, which use AI to improve their operations and outpace competitors. At Alphabet subsidiary Google, for example, AI is central to its eponymous search engine, and self-driving car company Waymo began as an Alphabet division. The Google Brain research lab also invented the transformer architecture that underpins recent NLP breakthroughs such as OpenAI’s ChatGPT.

Types of Artificial Intelligence

Artificial intelligence can be classified in several different ways.

Strong AI vs. Weak AI

AI can be organized into two broad categories: weak AI and strong AI.

Weak AI (or narrow AI) refers to AI that automates specific tasks. It typically outperforms humans, but it operates within a limited context and is applied to a narrowly defined problem. For now, all AI systems are examples of weak AI, ranging from email inbox spam filters to recommendation engines to chatbots.
Strong AI, often referred to as artificial general intelligence (AGI), is a hypothetical benchmark at which AI could possess human-like intelligence and adaptability, solving problems it’s never been trained to work on. AGI does not actually exist yet, and it is unclear whether it ever will.

The 4 Kinds of AI

AI can then be further categorized into four main types: reactive machines, limited memory, theory of mind and self-awareness.

Reactive machines perceive the world in front of them and react. They can carry out specific commands and requests, but they cannot store memory or rely on past experiences to inform their decision making in real time. This makes reactive machines useful for completing a limited number of specialized duties. Examples include Netflix’s recommendation engine and IBM’s Deep Blue (used to play chess).
Limited memory AI has the ability to store previous data and predictions when gathering information and making decisions. Essentially, it looks into the past for clues to predict what may come next. Limited memory AI is created when a team continuously trains a model in how to analyze and utilize new data, or an AI environment is built so models can be automatically trained and renewed. Examples include ChatGPT and self-driving cars.
Theory of mind is a type of AI that does not actually exist yet, but it describes the idea of an AI system that can perceive and understand human emotions, and then use that information to predict future actions and make decisions on its own.
Self-aware AI refers to artificial intelligence that has self-awareness, or a sense of self. This type of AI does not currently exist. In theory, though, self-aware AI possesses human-like consciousness and understands its own existence in the world, as well as the emotional state of others.

Advantages of AI

The following are some advantages of AI:

Excellence in detail-oriented jobs. AI is a good fit for tasks that involve identifying subtle patterns and relationships in data that might be overlooked by humans. For example, in oncology, AI systems have demonstrated high accuracy in detecting early-stage cancers, such as breast cancer and melanoma, by highlighting areas of concern for further evaluation by healthcare professionals.
Efficiency in data-heavy tasks. AI systems and automation tools dramatically reduce the time required for data processing. This is particularly useful in sectors like finance, insurance and healthcare that involve a great deal of routine data entry and analysis, as well as data-driven decision-making. For example, in banking and finance, predictive AI models can process vast volumes of data to forecast market trends and analyze investment risk.
Time savings and productivity gains. AI and robotics can not only automate operations but also improve safety and efficiency. In manufacturing, for example, AI-powered robots are increasingly used to perform hazardous or repetitive tasks as part of warehouse automation, thus reducing the risk to human workers and increasing overall productivity.
Consistency in results. Today’s analytics tools use AI and machine learning to process extensive amounts of data in a uniform way, while retaining the ability to adapt to new information through continuous learning. For example, AI applications have delivered consistent and reliable outcomes in legal document review and language translation.
Customization and personalization. AI systems can enhance user experience by personalizing interactions and content delivery on digital platforms. On e-commerce platforms, for example, AI models analyze user behavior to recommend products suited to an individual’s preferences, increasing customer satisfaction and engagement.
Round-the-clock availability. AI programs do not need to sleep or take breaks. For example, AI-powered virtual assistants can provide uninterrupted, 24/7 customer service even under high interaction volumes, improving response times and reducing costs.
Scalability. AI systems can scale to handle growing amounts of work and data. This makes AI well suited for scenarios where data volumes and workloads can grow exponentially, such as internet search and business analytics.
Accelerated research and development. AI can speed up the pace of R&D in fields such as pharmaceuticals and materials science. By rapidly simulating and analyzing many possible scenarios, AI models can help researchers discover new drugs, materials or compounds more quickly than traditional methods.
Sustainability and conservation. AI and machine learning are increasingly used to monitor environmental changes, predict future weather events and manage conservation efforts. Machine learning models can process satellite imagery and sensor data to track wildfire risk, pollution levels and endangered species populations, for example.
Process optimization. AI is used to streamline and automate complex processes across various industries. For example, AI models can identify inefficiencies and predict bottlenecks in manufacturing workflows, while in the energy sector, they can forecast electricity demand and allocate supply in real time.

Disadvantages of AI

The following are some disadvantages of AI:

High costs. Developing AI can be very expensive. Building an AI model requires a substantial upfront investment in infrastructure, computational resources and software to train the model and store its training data. After initial training, there are further ongoing costs associated with model inference and retraining. As a result, costs can rack up quickly, particularly for advanced, complex systems like generative AI applications; OpenAI CEO Sam Altman has stated that training the company’s GPT-4 model cost over $100 million.
Technical complexity. Developing, operating and troubleshooting AI systems — especially in real-world production environments — requires a great deal of technical know-how. In many cases, this knowledge differs from that needed to build non-AI software. For example, building and deploying a machine learning application involves a complex, multistage and highly technical process, from data preparation to algorithm selection to parameter tuning and model testing.
Talent gap. Compounding the problem of technical complexity, there is a significant shortage of professionals trained in AI and machine learning compared with the growing need for such skills. This gap between AI talent supply and demand means that, even though interest in AI applications is growing, many organizations cannot find enough qualified workers to staff their AI initiatives.
Algorithmic bias. AI and machine learning algorithms reflect the biases present in their training data — and when AI systems are deployed at scale, the biases scale, too. In some cases, AI systems may even amplify subtle biases in their training data by encoding them into reinforceable and pseudo-objective patterns. In one well-known example, Amazon developed an AI-driven recruitment tool to automate the hiring process that inadvertently favored male candidates, reflecting larger-scale gender imbalances in the tech industry.
Difficulty with generalization. AI models often excel at the specific tasks for which they were trained but struggle when asked to address novel scenarios. This lack of flexibility can limit AI’s usefulness, as new tasks might require the development of an entirely new model. An NLP model trained on English-language text, for example, might perform poorly on text in other languages without extensive additional training. While work is underway to improve models’ generalization ability — known as domain adaptation or transfer learning — this remains an open research problem.

Job displacement. AI can lead to job loss if organizations replace human workers with machines — a growing area of concern as the capabilities of AI models become more sophisticated and companies increasingly look to automate workflows using AI. For example, some copywriters have reported being replaced by large language models (LLMs) such as ChatGPT. While widespread AI adoption may also create new job categories, these may not overlap with the jobs eliminated, raising concerns about economic inequality and reskilling.
Security vulnerabilities. AI systems are susceptible to a wide range of cyberthreats, including data poisoning and adversarial machine learning. Hackers can extract sensitive training data from an AI model, for example, or trick AI systems into producing incorrect and harmful output. This is particularly concerning in security-sensitive sectors such as financial services and government.
Environmental impact. The data centers and network infrastructures that underpin the operations of AI models consume large amounts of energy and water. Consequently, training and running AI models has a significant impact on the climate. AI’s carbon footprint is especially concerning for large generative models, which require a great deal of computing resources for training and ongoing use.
Legal issues. AI raises complex questions around privacy and legal liability, particularly amid an evolving AI regulation landscape that differs across regions. Using AI to analyze and make decisions based on personal data has serious privacy implications, for example, and it remains unclear how courts will view the authorship of material generated by LLMs trained on copyrighted works.

What are the applications of AI?

AI has entered a wide variety of industry sectors and research areas. The following are several of the most notable examples.

AI in healthcare

AI is applied to a range of tasks in the healthcare domain, with the overarching goals of improving patient outcomes and reducing systemic costs. One major application is the use of machine learning models trained on large medical data sets to assist healthcare professionals in making better and faster diagnoses. For example, AI-powered software can analyze CT scans and alert neurologists to suspected strokes.

On the patient side, online virtual health assistants and chatbots can provide general medical information, schedule appointments, explain billing processes and complete other administrative tasks. Predictive modeling AI algorithms can also be used to combat the spread of pandemics such as COVID-19.

AI in business

AI is increasingly integrated into various business functions and industries, aiming to improve efficiency, customer experience, strategic planning and decision-making. For example, machine learning models power many of today’s data analytics and customer relationship management (CRM) platforms, helping companies understand how to best serve customers through personalizing offerings and delivering better-tailored marketing.

Virtual assistants and chatbots are also deployed on corporate websites and in mobile applications to provide round-the-clock customer service and answer common questions. In addition, more and more companies are exploring the capabilities of generative AI tools such as ChatGPT for automating tasks such as document drafting and summarization, product design and ideation, and computer programming.

AI in education

AI has a number of potential applications in education technology. It can automate aspects of grading processes, giving educators more time for other tasks. AI tools can also assess students’ performance and adapt to their individual needs, facilitating more personalized learning experiences that enable students to work at their own pace. AI tutors could also provide additional support to students, ensuring they stay on track. The technology could also change where and how students learn, perhaps altering the traditional role of educators.

As the capabilities of LLMs such as ChatGPT and Google Gemini grow, such tools could help educators craft teaching materials and engage students in new ways. However, the advent of these tools also forces educators to reconsider homework and testing practices and revise plagiarism policies, especially given that AI detection and AI watermarking tools are currently unreliable.

AI in finance and banking

Banks and other financial organizations use AI to improve their decision-making for tasks such as granting loans, setting credit limits and identifying investment opportunities. In addition, algorithmic trading powered by advanced AI and machine learning has transformed financial markets, executing trades at speeds and efficiencies far surpassing what human traders could do manually.

AI and machine learning have also entered the realm of consumer finance. For example, banks use AI chatbots to inform customers about services and offerings and to handle transactions and questions that don’t require human intervention. Similarly, Intuit offers generative AI features within its TurboTax e-filing product that provide users with personalized advice based on data such as the user’s tax profile and the tax code for their location.

AI in law

AI is changing the legal sector by automating labor-intensive tasks such as document review and discovery response, which can be tedious and time consuming for attorneys and paralegals. Law firms today use AI and machine learning for a variety of tasks, including analytics and predictive AI to analyze data and case law, computer vision to classify and extract information from documents, and NLP to interpret and respond to discovery requests.

In addition to improving efficiency and productivity, this integration of AI frees up human legal professionals to spend more time with clients and focus on more creative, strategic work that AI is less well suited to handle. With the rise of generative AI in law, firms are also exploring using LLMs to draft common documents, such as boilerplate contracts.

AI in entertainment and media

The entertainment and media business uses AI techniques in targeted advertising, content recommendations, distribution and fraud detection. The technology enables companies to personalize audience members’ experiences and optimize delivery of content.

Generative AI is also a hot topic in the area of content creation. Advertising professionals are already using these tools to create marketing collateral and edit advertising images. However, their use is more controversial in areas such as film and TV scriptwriting and visual effects, where they offer increased efficiency but also threaten the livelihoods and intellectual property of humans in creative roles.

AI in journalism

In journalism, AI can streamline workflows by automating routine tasks, such as data entry and proofreading. Investigative journalists and data journalists also use AI to find and research stories by sifting through large data sets using machine learning models, thereby uncovering trends and hidden connections that would be time consuming to identify manually. For example, five finalists for the 2024 Pulitzer Prizes for journalism disclosed using AI in their reporting to perform tasks such as analyzing massive volumes of police records. While the use of traditional AI tools is increasingly common, the use of generative AI to write journalistic content is open to question, as it raises concerns around reliability, accuracy and ethics.

AI in software development and IT

AI is used to automate many processes in software development, DevOps and IT. For example, AIOps tools enable predictive maintenance of IT environments by analyzing system data to forecast potential issues before they occur, and AI-powered monitoring tools can help flag potential anomalies in real time based on historical system data. Generative AI tools such as GitHub Copilot and Tab nine are also increasingly used to produce application code based on natural-language prompts. While these tools have shown early promise and interest among developers, they are unlikely to fully replace software engineers. Instead, they serve as useful productivity aids, automating repetitive tasks and boilerplate code writing.

AI in security

AI and machine learning are prominent buzzwords in security vendor marketing, so buyers should take a cautious approach. Still, AI is indeed a useful technology in multiple aspects of cybersecurity, including anomaly detection, reducing false positives and conducting behavioral threat analytics. For example, organizations use machine learning in security information and event management (SIEM) software to detect suspicious activity and potential threats. By analyzing vast amounts of data and recognizing patterns that resemble known malicious code, AI tools can alert security teams to new and emerging attacks, often much sooner than human employees and previous technologies could.

AI in manufacturing

Manufacturing has been at the forefront of incorporating robots into workflows, with recent advancements focusing on collaborative robots, or cobots. Unlike traditional industrial robots, which were programmed to perform single tasks and operated separately from human workers, cobots are smaller, more versatile and designed to work alongside humans. These multitasking robots can take on responsibility for more tasks in warehouses, on factory floors and in other workspaces, including assembly, packaging and quality control. In particular, using robots to perform or assist with repetitive and physically demanding tasks can improve safety and efficiency for human workers.

AI in transportation

In addition to AI’s fundamental role in operating autonomous vehicles, AI technologies are used in automotive transportation to manage traffic, reduce congestion and enhance road safety. In air travel, AI can predict flight delays by analyzing data points such as weather and air traffic conditions. In overseas shipping, AI can enhance safety and efficiency by optimizing routes and automatically monitoring vessel conditions.

In supply chains, AI is replacing traditional methods of demand forecasting and improving the accuracy of predictions about potential disruptions and bottlenecks. The COVID-19 pandemic highlighted the importance of these capabilities, as many companies were caught off guard by the effects of a global pandemic on the supply and demand of goods.