Lord777
Professional
- Messages
- 2,579
- Reaction score
- 1,513
- Points
- 113
Artificial intelligence (AI) is the modeling of human brain processes by computer systems. AI includes expert systems, natural language processing, speech recognition, and machine vision.
How does artificial intelligence work?
AI requires specialized hardware and software to write and train algorithms. For AI, the programming languages Python, R, and Java are mainly used.
As a rule, the work of AI consists in processing large amounts of training data in order to analyze them for correlations and patterns. These patterns are then used to predict future states. Thus, a chatbot that receives examples of text chats can learn how to produce realistic dialogue with people, and an image recognition tool can learn how to identify and describe objects in images by studying millions of examples.
AI programming focuses on three cognitive skills: learning, reasoning, and self-correction.
Why is artificial intelligence important?
AI can give an organization insight into its activities that it didn't know about before. AI is also important because in some cases AI can perform tasks better than humans. In particular, when it comes to repetitive, detail-oriented tasks, such as analyzing a large number of legal documents to correctly fill in the relevant fields, AI tools often perform the job quickly and with relatively few errors.
This has led to skyrocketing productivity growth and opened the door to entirely new opportunities for some large businesses.
Before the rise of AI, it was hard to imagine using computer software to connect passengers to taxis, but today Uber has become one of the largest companies in the world by doing just that. Uber uses sophisticated Machine Learning (ML) algorithms to predict when people will need a taxi in certain areas. This allows you to get drivers to the right area in advance before they are needed.
In addition, Google has become one of the largest players in the online services market, using machine learning to analyze how people use the company's services, and then improving them. Back in 2017, the company's CEO, Sundar Pichai, said that Google would operate as a company "primarily engaged in artificial intelligence."
Today, the largest and most successful businesses use AI to improve their operations and gain an advantage over their competitors.
Advantages and disadvantages of artificial intelligence
Neural networks and deep learning technologies are rapidly developing primarily because AI processes large amounts of data much faster and makes predictions more accurate than humans.
The main disadvantage of using AI is the high cost of processing large amounts of data required for programming artificial intelligence.
Advantages
Disadvantages
Strong and weak AI
AI can be divided into weak and strong.
4 types of artificial intelligence
Arend Hintze, an assistant professor of integrative biology, computer science and engineering at the University of Michigan, explained in his article that AI can be divided into 4 types:
How is artificial intelligence used today?
AI is used in many different types of technologies. Here are 6 examples:
What applications are there based on artificial intelligence?
Artificial intelligence has penetrated into a variety of niches. Here are 9 examples.
AI in healthcare. The biggest bets are placed on improving patient outcomes and reducing costs. Companies use machine learning to make better and faster diagnoses than people do.
One of the most well-known technologies used in healthcare is IBM Watson technology. The model understands natural language and can answer the questions asked. The system analyzes patient data and other available data sources to form a hypothesis, which it then presents using a confidence assessment scheme.
Other AI applications include the use of virtual medical assistants and chatbots that help patients and clients in the healthcare industry find medical information, schedule doctor appointments, understand the billing process, and perform other administrative processes. A host of AI technologies are also being used to predict, manage, and better understand pandemics, as was the case with COVID-19.
AI in business. Machine learning algorithms are integrated into analytics platforms and CRM systems to provide information on how to better serve customers. In addition, chatbots have long been built into websites for instant customer service. Workplace automation has also become a topic of discussion among scientists and IT analysts.
AI in education. AI can automate grading, as well as evaluate students and adapt to their needs, helping them work at their own pace.
Virtual curators can provide additional support for students to keep up with the curriculum. Perhaps AI technologies can change the approach to learning by replacing some teachers.
AI in the financial sector. AI in personal finance and tax filing apps is changing the way financial institutions work. These apps collect personal data from customers and provide them with financial advice. In addition, at the moment, AI is used in the process of buying real estate and performs most of the trading on Wall Street.
AI in law. The process of selecting documents in law is often difficult for people. Using AI to automate time-consuming processes in the legal industry saves time and improves customer service. Law firms use machine learning to describe data and predict results, computer vision to classify and extract information from documents, and natural language processing to interpret requests for information.
AI in production. Industrial robots that were once programmed to perform separate tasks and separate from humans are increasingly functioning as cobots: small multitasking robots that collaborate with humans and perform a large number of operations in warehouses, factory floors, and other workplaces.
AI in the banking sector. Banks successfully use chatbots to inform customers about services and offers, as well as to process transactions that do not require human intervention. Banks also use AI to improve their loan decision-making process, as well as to set credit limits and identify investment opportunities.
AI in the field of transport services. In addition to the fundamental role of AI in the management of autonomous vehicles, AI technologies are used in the field of transport services to manage traffic, predict flight delays, and improve the safety and efficiency of maritime transport.
Safety. Organizations use machine learning in SIEM (Security Event Management System) and related areas to detect anomalies and detect suspicious activity on the network. By analyzing data and using logic to identify similarities with known malicious code, AI can warn about new and emerging attacks much earlier than" live " employees. Emerging technologies play an important role in the fight against cyber attacks.
Augmented Intelligence vs Artificial Intelligence
Some industry experts believe that the term "artificial intelligence" is too closely associated with popular culture, and this has led to incredible expectations among the general public about how AI will change the workplace and life in general.
Ethical use of Artificial intelligence
Artificial intelligence is also controversial about the ethics of using it – for better or worse, an AI system remembers what it has already learned.
The machine learning algorithms that underpin many of the most advanced AI tools are only as smart as their training data is good. Since the training data is chosen by humans, the potential shift in machine learning is inevitable and must be carefully monitored.
When using machine learning as part of real production systems, you need to take into account the issue of ethics in AI training processes and avoid bias. This is especially important when using Deep Learning and generative-adversarial network (GAN) algorithms.
The ability to explain a decision is a big challenge when using AI in industries where there are strict regulatory requirements. For example, financial institutions in the United States must explain their decision to grant or refuse a loan. But when the decision to reject is made by a neural network, it is difficult to explain its reasons, because AI tools reveal subtle correlations between thousands of variables. When the decision-making process cannot be justified, an AI program is called a "black box".
Despite the potential risks, there are currently several rules governing the use of AI tools. The EU's General Data Protection Regulation (GDPR) places strict restrictions on how businesses can use user data, which hinders the learning and functionality of many consumer-facing AI applications.
Developing laws to regulate AI is not an easy task, partly because AI involves a variety of technologies that companies use for different purposes, and also because regulation can be achieved through the progress and development of AI.
The rapid evolution of AI technologies is another obstacle to the formation of an AI regulatory mechanism. A technological breakthrough can instantly make existing laws obsolete.
For example, the laws governing the privacy of recorded conversations do not apply to voice assistants (Amazon's Alexa and Apple's Siri). They collect but do not distribute conversations, except for company technicians who use recordings of conversations with the assistant to improve algorithms.
In addition, the laws that governments develop to regulate AI do not prevent criminals from using this technology with malicious intent.
Cognitive Computing and AI
The terms "artificial intelligence" and "cognitive computing" are sometimes used interchangeably, but in general, the term "artificial intelligence" is used to refer to machines that replace human intelligence by mimicking how we learn, perceive, process, and respond to information in the real world.
The term "cognitive computing" is used to refer to products and services that mimic and complement human thought processes.
What is the history of AI?
The idea of inanimate objects endowed with intelligence has existed since ancient times. The Greek god Hephaestus was depicted in myth as forging robot-like servants out of gold. Engineers in ancient Egypt built statues of gods animated by priests.
For centuries, thinkers from Aristotle to the 13th-century Spanish theologian Ramon Lullia, Rene Descartes, and Thomas Bayes have used the tools and logic of their time to describe human thought processes as symbols, laying the groundwork for AI concepts such as the representation of publicly available knowledge.
AI as a service
Because the costs of AI maintenance hardware, software, and personnel can be high, many vendors include AI components in their standard offerings or provide access to artificial intelligence as a service (AIaaS) platforms.
AIaaS is a cloud-based service that offers businesses artificial intelligence and machine learning capabilities. AIaaS provides ready-made models and platforms for natural language processing, computer vision, speech recognition, predictive analytics, and other machine learning capabilities.
AIaaS reduces costs, increases scalability, and allows businesses to focus on developing innovative solutions. Examples of AIaaS providers include Google Cloud AI Platform, AWS AI, Microsoft Azure Machine Learning, and IBM Watson Studio.
How does artificial intelligence work?
AI requires specialized hardware and software to write and train algorithms. For AI, the programming languages Python, R, and Java are mainly used.
As a rule, the work of AI consists in processing large amounts of training data in order to analyze them for correlations and patterns. These patterns are then used to predict future states. Thus, a chatbot that receives examples of text chats can learn how to produce realistic dialogue with people, and an image recognition tool can learn how to identify and describe objects in images by studying millions of examples.
AI programming focuses on three cognitive skills: learning, reasoning, and self-correction.
- Learning processes. Collect data and create algorithms (rules) for turning data into useful information. Algorithms provide machines with step-by-step instructions on how to perform a specific task;
- Processes of reasoning. Choosing the right algorithm to achieve the desired result;
- Self-correction processes. Constant fine-tuning of algorithms and ensuring the most accurate results.
Why is artificial intelligence important?
AI can give an organization insight into its activities that it didn't know about before. AI is also important because in some cases AI can perform tasks better than humans. In particular, when it comes to repetitive, detail-oriented tasks, such as analyzing a large number of legal documents to correctly fill in the relevant fields, AI tools often perform the job quickly and with relatively few errors.
This has led to skyrocketing productivity growth and opened the door to entirely new opportunities for some large businesses.
Before the rise of AI, it was hard to imagine using computer software to connect passengers to taxis, but today Uber has become one of the largest companies in the world by doing just that. Uber uses sophisticated Machine Learning (ML) algorithms to predict when people will need a taxi in certain areas. This allows you to get drivers to the right area in advance before they are needed.
In addition, Google has become one of the largest players in the online services market, using machine learning to analyze how people use the company's services, and then improving them. Back in 2017, the company's CEO, Sundar Pichai, said that Google would operate as a company "primarily engaged in artificial intelligence."
Today, the largest and most successful businesses use AI to improve their operations and gain an advantage over their competitors.
Advantages and disadvantages of artificial intelligence
Neural networks and deep learning technologies are rapidly developing primarily because AI processes large amounts of data much faster and makes predictions more accurate than humans.
The main disadvantage of using AI is the high cost of processing large amounts of data required for programming artificial intelligence.
Advantages
- It does a good job that requires attention to detail;
- Reduces the time required for solving problems with a large amount of data;
- Provides consistent results;
- AI-based virtual agents are always available.
Disadvantages
- Expensive technology;
- Requires deep technical knowledge;
- Limited number of qualified specialists to create AI tools;
- He only knows what he has learned in the course of training;
- Lack of ability to generalize tasks.
Strong and weak AI
AI can be divided into weak and strong.
- Weak (limited) AI is an artificial intelligence designed and trained to perform a specific task. Industrial robots and virtual assistants (such as Siri and Cortana) use weak AI.
- Strong (general) AI is an artificial intelligence that can replicate the cognitive abilities of the human brain. When solving an unfamiliar problem, a strong AI can use fuzzy logic to apply knowledge from one area to another and independently search for a solution to the problem. A strong AI must pass both the Turing test and the Chinese room test, designed to test the" thinking " of machines.
4 types of artificial intelligence
Arend Hintze, an assistant professor of integrative biology, computer science and engineering at the University of Michigan, explained in his article that AI can be divided into 4 types:
- Jet engines. These AI systems have no memory and are task-specific. An example is Deep Blue, the IBM chess program that beat Garry Kasparov in the 1990s.
- Deep Blue can identify pieces on a chessboard and make predictions, but because it has no memory, it can't use past experience to justify future decisions.
- Suitable for simple image classification and recognition tasks;
- Suitable for scenarios where all parameters are known: it can outperform humans because it can perform calculations much faster;
- Inability to work with scenarios that involve imperfect information or require historical understanding.
- Limited memory. These AI systems have memory, so they can use past experience to inform future decisions. Some decision-making functions in self-driving cars are designed in this way.
- It can solve complex classification problems;
- Uses historical data for forecasting;
- It can perform complex tasks, such as driving autonomous cars, but still can't handle extraneous values or negative examples.;
- Existing AI systems are of this type, and some experts say that we have "hit a wall" in terms of AI development.
- Theory of consciousness. Theory of consciousness is a term from psychology. When applied to AI, this means that the system must have social intelligence in order to understand emotions. This type of AI will be able to guess human intentions and predict behavior – a necessary skill for artificial intelligence, which can become an integral member of human teams.
- Able to understand human motivations and reasoning. Can provide a personal experience to everyone based on their motivations and needs;
- He is able to learn from fewer examples, because he understands motives and intentions;
- It is considered the next milestone in the evolution of AI.
- Self-awareness. In this category, AI systems have a" sense of self" that gives them consciousness. This is a human-level artificial intelligence that can also bypass human intelligence. Self-aware machines understand their current state. This type of AI does not currently exist.
How is artificial intelligence used today?
AI is used in many different types of technologies. Here are 6 examples:
- Automation. When combined with AI technologies, automation tools can expand the scope and types of tasks performed. An example is robotic process automation (RPA), a type of software that automates repetitive rule – based data processing tasks traditionally performed by humans.
- Combined with machine learning and new AI tools, RPA can automate much of the work of enterprises, allowing operational RPA bots to transmit information from AI and respond to process changes.
- Machine learning.This is the science of how to make a computer work without programming. Deep learning is a subsection of machine learning that can be considered as automation of predictive analytics.
- There are three types of machine learning algorithms:
- Machine learning with a teacher. Data sets are labeled so that patterns can be detected and used to label new data sets.;
- Unsupervised learning. Data sets are not labeled and are sorted by similarity or difference;
- Learning with reinforcement signals from the interaction environment. Data sets are not labeled, but after performing an action or several actions, the AI system receives feedback.
- Machine vision. This technology allows the machine to see. Machine vision captures and analyzes visual information using a camera, analog-to-digital conversion, and digital signal processing. Machine vision is often compared to human vision, but it can be programmed, for example, to see through walls.
- Machine vision is used in a range of applications from signature identification to medical image analysis. Computer vision focused on machine image processing is often identified with machine vision.
- Natural Language Information Processing (NLP). This is the processing of human language by a computer program. One of the oldest and most well — known examples of NLP is spam detection, which looks at the subject line and text of an email and decides whether it is spam. Modern approaches to NLP are based on machine learning. NLP tasks include text translation, sentiment analysis, and speech recognition.
- Robotization. This engineering field deals with the design and production of robots. Robots are often used to perform tasks that are difficult for humans to perform or to perform tasks sequentially. For example, robots are used on assembly lines to produce cars or by NASA to move large objects in space. Researchers are also using machine learning to create robots that can interact on social media.
- Self-driving cars. Autonomous vehicles use a combination of computer vision, image recognition, and deep learning to develop automated vehicle piloting skills while staying in a given lane and avoiding unexpected obstacles and pedestrians.
What applications are there based on artificial intelligence?
Artificial intelligence has penetrated into a variety of niches. Here are 9 examples.
AI in healthcare. The biggest bets are placed on improving patient outcomes and reducing costs. Companies use machine learning to make better and faster diagnoses than people do.
One of the most well-known technologies used in healthcare is IBM Watson technology. The model understands natural language and can answer the questions asked. The system analyzes patient data and other available data sources to form a hypothesis, which it then presents using a confidence assessment scheme.
Other AI applications include the use of virtual medical assistants and chatbots that help patients and clients in the healthcare industry find medical information, schedule doctor appointments, understand the billing process, and perform other administrative processes. A host of AI technologies are also being used to predict, manage, and better understand pandemics, as was the case with COVID-19.
AI in business. Machine learning algorithms are integrated into analytics platforms and CRM systems to provide information on how to better serve customers. In addition, chatbots have long been built into websites for instant customer service. Workplace automation has also become a topic of discussion among scientists and IT analysts.
AI in education. AI can automate grading, as well as evaluate students and adapt to their needs, helping them work at their own pace.
Virtual curators can provide additional support for students to keep up with the curriculum. Perhaps AI technologies can change the approach to learning by replacing some teachers.
AI in the financial sector. AI in personal finance and tax filing apps is changing the way financial institutions work. These apps collect personal data from customers and provide them with financial advice. In addition, at the moment, AI is used in the process of buying real estate and performs most of the trading on Wall Street.
AI in law. The process of selecting documents in law is often difficult for people. Using AI to automate time-consuming processes in the legal industry saves time and improves customer service. Law firms use machine learning to describe data and predict results, computer vision to classify and extract information from documents, and natural language processing to interpret requests for information.
AI in production. Industrial robots that were once programmed to perform separate tasks and separate from humans are increasingly functioning as cobots: small multitasking robots that collaborate with humans and perform a large number of operations in warehouses, factory floors, and other workplaces.
AI in the banking sector. Banks successfully use chatbots to inform customers about services and offers, as well as to process transactions that do not require human intervention. Banks also use AI to improve their loan decision-making process, as well as to set credit limits and identify investment opportunities.
AI in the field of transport services. In addition to the fundamental role of AI in the management of autonomous vehicles, AI technologies are used in the field of transport services to manage traffic, predict flight delays, and improve the safety and efficiency of maritime transport.
Safety. Organizations use machine learning in SIEM (Security Event Management System) and related areas to detect anomalies and detect suspicious activity on the network. By analyzing data and using logic to identify similarities with known malicious code, AI can warn about new and emerging attacks much earlier than" live " employees. Emerging technologies play an important role in the fight against cyber attacks.
Augmented Intelligence vs Artificial Intelligence
Some industry experts believe that the term "artificial intelligence" is too closely associated with popular culture, and this has led to incredible expectations among the general public about how AI will change the workplace and life in general.
- Augmented intelligence. Some researchers and marketers hope that the label "augmented intelligence", which has a more neutral connotation, will help people understand that most AI developments will be weak and simply improve products and services. This can include, for example, automatically displaying important information in business intelligence reports or highlighting important information in legal documents.
- Artificial intelligence. True (general) artificial intelligence is closely related to the concept of a technological singularity — a future driven by an artificial superintelligence that far exceeds the human brain's ability to understand it or how it shapes our reality.
Ethical use of Artificial intelligence
Artificial intelligence is also controversial about the ethics of using it – for better or worse, an AI system remembers what it has already learned.
The machine learning algorithms that underpin many of the most advanced AI tools are only as smart as their training data is good. Since the training data is chosen by humans, the potential shift in machine learning is inevitable and must be carefully monitored.
When using machine learning as part of real production systems, you need to take into account the issue of ethics in AI training processes and avoid bias. This is especially important when using Deep Learning and generative-adversarial network (GAN) algorithms.
The ability to explain a decision is a big challenge when using AI in industries where there are strict regulatory requirements. For example, financial institutions in the United States must explain their decision to grant or refuse a loan. But when the decision to reject is made by a neural network, it is difficult to explain its reasons, because AI tools reveal subtle correlations between thousands of variables. When the decision-making process cannot be justified, an AI program is called a "black box".
Despite the potential risks, there are currently several rules governing the use of AI tools. The EU's General Data Protection Regulation (GDPR) places strict restrictions on how businesses can use user data, which hinders the learning and functionality of many consumer-facing AI applications.
Developing laws to regulate AI is not an easy task, partly because AI involves a variety of technologies that companies use for different purposes, and also because regulation can be achieved through the progress and development of AI.
The rapid evolution of AI technologies is another obstacle to the formation of an AI regulatory mechanism. A technological breakthrough can instantly make existing laws obsolete.
For example, the laws governing the privacy of recorded conversations do not apply to voice assistants (Amazon's Alexa and Apple's Siri). They collect but do not distribute conversations, except for company technicians who use recordings of conversations with the assistant to improve algorithms.
In addition, the laws that governments develop to regulate AI do not prevent criminals from using this technology with malicious intent.
Cognitive Computing and AI
The terms "artificial intelligence" and "cognitive computing" are sometimes used interchangeably, but in general, the term "artificial intelligence" is used to refer to machines that replace human intelligence by mimicking how we learn, perceive, process, and respond to information in the real world.
The term "cognitive computing" is used to refer to products and services that mimic and complement human thought processes.
What is the history of AI?
The idea of inanimate objects endowed with intelligence has existed since ancient times. The Greek god Hephaestus was depicted in myth as forging robot-like servants out of gold. Engineers in ancient Egypt built statues of gods animated by priests.
For centuries, thinkers from Aristotle to the 13th-century Spanish theologian Ramon Lullia, Rene Descartes, and Thomas Bayes have used the tools and logic of their time to describe human thought processes as symbols, laying the groundwork for AI concepts such as the representation of publicly available knowledge.
AI as a service
Because the costs of AI maintenance hardware, software, and personnel can be high, many vendors include AI components in their standard offerings or provide access to artificial intelligence as a service (AIaaS) platforms.
AIaaS is a cloud-based service that offers businesses artificial intelligence and machine learning capabilities. AIaaS provides ready-made models and platforms for natural language processing, computer vision, speech recognition, predictive analytics, and other machine learning capabilities.
AIaaS reduces costs, increases scalability, and allows businesses to focus on developing innovative solutions. Examples of AIaaS providers include Google Cloud AI Platform, AWS AI, Microsoft Azure Machine Learning, and IBM Watson Studio.
