Introduction
    Is artificial intelligence progress and new opportunities, or is it a challenge for humans and a curse for society?
    In the last decade, there has been a growing interest in the use of artificial intelligence in all spheres of life. AI has become an integral part of many disciplines over the years, including natural sciences. It changes not only the approaches to conducting research, but also has a profound impact on the very essence of the scientific process. Artificial intelligence is transforming modern science and opening up new horizons for scientific research.
Now we not only talk to voice assistants, but also see AI in human work at various levels — from solving everyday tasks to direct support in scientific research. AI has a special impact on the field of science and medicine, significantly expanding the capabilities and achievements of researchers. Let's look at how AI affects scientific and medical research, offering new perspectives and opportunities, but at the same time creating problems that require attention and developing solutions to optimize further collaboration. 
The relevance of the work lies in the fact that AI opens up new opportunities and raises questions about its impact on scientific research and medicine. The topic of AI application in science and medicine is relevant and significant, as it is able to automate and optimize research processes, as well as introduce new learning algorithms and create new opportunities for the development of science and medicine.
The purpose of this article is to consider new opportunities and challenges related to the use of AI in scientific research and medicine. To achieve this goal, the article evaluates the use of AI, examines the limitations of its use, and explores the ethical, legal and social aspects of its application. In addition, the article explores the role of AI learning algorithms and the formation of new opportunities for the development of science and medicine.
As part of the fulfillment of these goals, the prospects for the use of AI and the new opportunities that it opens up for scientific research and medicine are discussed. The implementation of forecasting in the field of AI application in scientific research and medicine will allow not only to conduct research more effectively, but also to introduce innovations and changes in modern practices. The purpose of forecasting is to determine the directions for further development of research, as well as to identify possible problems and challenges associated with the use of AI.

 
Artificial intelligence and modern technologies
    So how has artificial intelligence gained such momentum over the past 10 years that it has become such a significant place in human life? Where did he come from and for what? 
    An artificial intelligence (AI) system is a software system that simulates the human thinking process on a computer [2]. At the same time, to create such a system, it is necessary to study not only the algorithm of this process and develop software tools that reproduce them on a computer, but also the human thinking process itself [1]. It turns out that AI methods involve a simple structural approach to the development of complex software decision-making systems and the creation of intelligent systems capable of performing tasks that usually require human intelligence.  Artificial intelligence is a field of computer science, the purpose of which is to develop hardware and software tools that allow a non-programmer user to set and solve their tasks [2].
AI has indeed made huge breakthroughs over the past 10 years, and this is due to several factors. Firstly, the emergence and development of powerful computers and computing technologies made it possible to process huge amounts of data and perform complex calculations in real time [1]. This made it possible to use more sophisticated algorithms and machine learning models, which significantly improved the results of AI. Secondly, the availability and volume of data began to increase significantly [1]. Large companies and organizations collect and store huge amounts of data that can be used to train artificial intelligence. This allows you to create more accurate and efficient AI models. 
In addition, the emergence of new machine learning algorithms, such as deep learning, has become a key factor in the development of AI [3]. Deep learning allows AI models to independently study and find patterns in data, which leads to higher accuracy and efficiency. Deep learning is training that allows computational models consisting of several levels of processing to study the representation of data with several levels of abstraction [3]. These methods have significantly improved the state of the art in speech recognition, visual recognition and object detection, and many other fields such as medicine and science. Deep learning discovers a complex structure in large datasets using a back propagation algorithm to indicate how the machine should change its internal parameters, which are used to calculate the representation at each level based on the previous one[4]. This AI training has led to breakthroughs in image processing, large amounts of data, and improved optimization in many areas. 
Let's look at the use of AI in medical practice. In November 2017, a robot named Xiaoyi became the first machine to pass a written test as part of the national medical licensing exam of China [5]. In addition, its developer iFlytek Co Ltd, a leading Chinese artificial intelligence company, stated that the robot scored 456 points, which is 96 points higher than the required grades (for completing the accreditation test, you could get a maximum of 600 points, and the passing minimum for passing is 360 points) [5]. The robot, whose algorithm is based on AI, automatically collected the necessary information and analyzed it, but at the same time the work was completed with 76% correct answers, which is a good result. In addition, this result demonstrates that there are still a number of limitations that can cause errors in execution, thereby reducing efficiency. Liu Qingfeng, chairman of iFlytek, said: "We will officially launch the robot in March 2018. It is not intended to replace doctors. Instead, it is designed to promote better collaboration between humans and machines in order to increase efficiency."[6] Hefei is conducting a pilot pilot project with a city hospital in China to see how such a robot can help doctors. 
The example shows that AI is not able to replace the work of a doctor, but at the same time it can analyze general information about the patient, then make an initial diagnosis based on the symptoms of the disease, biochemical blood and urine tests, X-rays. Artificial intelligence can also analyze various treatment options and suggest the optimal course based on the specific characteristics of the patient.

The use of AI in science and medicine
    There are 2 parts to the application of artificial intelligence: cybernetics and neurocybernetics [7]. Neurocybernetics is focused on the detailed study of hardware modeling of structures similar to the structure of the human brain. They use an AI algorithm similar to the work of human neurons, which is why such networks are called neural networks. Cybernetics, on the contrary, are aimed at studying not the internal structure of the device, but the final result and the reaction of the device-human connection. Such devices are called "black box" cybernetics. [7].
    At the moment, products using AI for medicine and science based on cybernetics and neurocybernetics are being created. AI is used in the field of drug development, genome research, protein analysis, disease diagnosis and this is just the beginning! 
AI can process a huge amount of medical data, including electronic medical records, laboratory results, and medical images. This allows doctors to quickly access information about patients and compare their situations with other cases, improving the quality of their decisions. The pharmaceutical industry is currently facing challenges in maintaining its drug development programs.[8] The drug development process based on feedback and inductive-deductive chain begins with existing results obtained from various sources, such as high-performance screening of compounds and fragments of genes, computer modeling, and information available in open sources [8]. AI processes a database set by specialists and leads to optimized data, such processes speed up work.
    The first step in drug development is the identification of new chemical compounds with biological activity. [8]. This biological activity can occur as a result of the interaction of the compound with specific enzymes. The first compound that exhibits activity against a given biological target is called "hit". Coincidences are often found when checking chemical libraries, as well as computer modeling or screening of naturally isolated drugs such as plants, bacteria and fungi.[9]. In the future, any chemical compound can be identified and analyzed. Such hits can be checked using AI in cytological analyses and further predict the behavior of a compound, cells or an organism (for example, bacteria). Scientists are expanding hit using different methods of organic chemistry. To increase synthesis performance, chemists focus on a specific reaction or a set of them in order to assemble the "building blocks", compounds with a functional group and an active center of a biological target, together and quickly obtain a series of analogues, which the AI, in turn, remembers and analyzes, learning from such reactions.[8]. 
 

What used to seem like fiction may become reality in the near future. 30 years ago, people would not have thought that robots could take part in operations. In the USA, the robot, named STAR (from Smart Tissue Autonomous Robot), was developed by Dr. Axel Krieger and his colleagues from Johns Hopkins University [12]. The robot was designed to perform intestinal anastomosis — when two pieces of small intestine tissue are stitched together to form a single continuous section — under the supervision and guidance of a surgeon.
    In medicine, AI is used as an auxiliary element not only in diagnostics and data analysis, but also as an element of cybernetics: for example, it significantly improves the accuracy of hardware diagnostics in medicine. Traditional diagnostic methods rely on human experience and interpretation of medical images, which can be subjective and error-prone. Artificial intelligence, on the other hand, can quickly and accurately analyze large amounts of medical data and images, resulting in a more accurate and effective diagnosis.
Artificial intelligence algorithms can be trained on extensive datasets of medical images, which allows them to study patterns and identify abnormalities that experts may find difficult to detect during an initial examination. This can be especially useful in the diagnosis of conditions such as cancer, cardiovascular diseases, and neurological disorders.[7]
AI-based diagnostic tools can also help healthcare professionals make more informed decisions by providing them with additional information and recommendations. For example, artificial intelligence algorithms can analyze electrocardiograms (ECGs) and help identify early signs of heart disease or predict the likelihood of a heart attack.

Limitations of the use of AI in the field of science and medicine
    The use of artificial intelligence in the field of medicine and science offers many opportunities and advantages, but there are certain limitations that should be taken into account.
1. Data quality: The use of AI requires the availability of high-quality and reliable data. If the data is incorrect, inaccurate, or incomplete, then all calculations and conclusions made by the AI may be unreliable and harmful.
2. Algorithmic transparency: Some AI algorithms are "black boxes" in which it is unclear how specific decisions are made. In medicine and science, it is important to be able to explain and interpret the decisions made, so the use of such algorithms can cause certain difficulties.
3. Responsibility: Making decisions based on AI data can raise questions about legal and ethical responsibility. Who is responsible for possible mistakes or incorrect conclusions made by AI? What precautions should be taken?
4. The human factor: In medicine and science, it is often necessary to take into account the human factor, such as intuition, professional experience and emotional perception. The use of AI does not completely replace the role of doctors and scientists, but it can be a useful tool to support decision-making. It can help in analyzing large amounts of data, searching for patterns and predicting results, which can significantly improve the quality of diagnosis and treatment. However, it is important to remember that the final decision should always be made by a doctor or scientist, taking into account all aspects and context of the patient or study.
5. Ethical considerations: The use of artificial intelligence raises ethical issues regarding data privacy, security, and bias. It is important to ensure that artificial intelligence algorithms are transparent, fair and do not cause any form of discrimination or inequality.
6. Technical limitations: Artificial intelligence systems may encounter technical failures or limitations that disrupt its learning process. [10]
7. Accessibility and access to technology: Some institutions may not have sufficient resources or experts to implement and use AI systems. In addition, online access to large amounts of data may be limited for privacy and cybersecurity reasons. On the one hand, these are serious shortcomings that limit work and progress, and on the other hand, this is a challenge for society, which can be solved by creating an open bank of scientific research and medical data of patients obtained through voluntary informed consent and consent to the processing of personal data.

Automation and optimization of research processes using AI
The use of artificial intelligence makes it possible to automate and optimize research processes in various fields of science. Here are some ways that AI can be useful:
1. Analyzing large amounts of data: AI allows you to process and analyze huge amounts of data, which allows scientists to discover hidden patterns, connections and relationships that would be difficult or impossible for conventional analysis methods. This helps scientists reduce the time and cost of research.
2. Predicting results: Using AI models and machine learning algorithms allows scientists to make predictions and predict the results of experiments. This is especially useful when it comes to complex systems or long-term studies where analytical prediction is difficult.
3. Process Automation: AI can automate routine tasks and processes such as data collection, processing, analysis, classification, or report generation. This frees up scientists' time for more creative work, such as developing new hypotheses or conducting experiments.
4. Finding new connections and improving research methodology: Using AI allows scientists to find new connections and statistical patterns in data faster.

Ethical, legal and social aspects of the use of AI in science and medicine
    There are several aspects of the use of AI in science and medicine that limit it. Let's start with the ethical issues.
1. Responsibility. All people make mistakes. And artificial intelligence created by humans can also make them, because AI is trained on databases and information created by humans. If a doctor makes a mistake, he is responsible for it. And if the AI makes a mistake and it leads to a deterioration of the patient's condition or to irreversible negative consequences that can cause death? Who is to blame? Is the attending physician responsible for erroneous decisions made based on AI recommendations? A car that made a mistake and misdiagnosed? The scientist who made the primary diagnostic database? Or even the patient himself, who agreed to it? This leads to the following problem.
2. Informed voluntary consent. If the influence of AI on healthcare and science increases in the future, then how to properly inform the patient that he will be "studied" by AI? The doctor will be obliged to clarify that AI will be involved directly in the treatment, because there must be transparency of the study. According to the law, the patient must be informed about the purposes of medical intervention, methods of providing medical care (including using AI), risks and consequences. Not every person will agree to entrust their life to a "soulless" machine. Many patients may refuse not only AI treatment, but also any medical care in general, due to a lack of understanding of the meaning and functions of such a device. Insufficient evidence of efficacy and safety, increased risk – all this can negatively affect treatment. 
3. Confidentiality and data protection. Using AI requires access to a large amount of data, including patient medical data and research information. It is necessary to ensure strict confidentiality and data protection in order to prevent possible violation of the privacy of patients and researchers. But what if AI needs to be trained for development, including on medical data and scientific research? 
4. The problem of the doctor's "black box" and the patient's fear of AI systems.  In the field of cybernetics, it is often difficult to explain the reason for the AI generated conclusions during training. For example, a doctor may not understand why the AI gives him this or that conclusion; the doctor will begin to treat him with distrust and eventually will not take into account the conclusion given by the system, and may even refuse to use AI systems at all. There is an opinion among some doctors that systems offering primary medical solutions and diagnostics can not only analyze electronic medical records and prompt doctors, but also report error statistics to higher authorities, thereby discrediting the work of a doctor in the eyes of a higher authority [13].  And this leads to a social problem related to the fact that in the future AI systems will be able to replace doctors, leaving them without work. But is it true?
And if we talk about the legal aspect of the use of AI, there are no less nuances. Due to the novelty and rapid progress in the field of AI, there are white spots in the legislation, that is, gaps in the law – the absence of a legal norm in resolving specific life cases that are covered by legal regulation and should be resolved on the basis of law [14]. 
To begin with, according to Russian laws, images and video files generated by neural networks are not objects of copyright [14]. They were created by a program that is not animated: that is, the results of its work formally do not belong to anyone. Technically, the legislation does not prohibit the use of this data even for commercial purposes. And then what to do with the data of medical and scientific research, in which there is confidentiality, but at the same time the AI is constantly learning from this data, processing and analyzing, thereby increasing its own database, gaining new "experience"? 

Returning to the problem of liability, from the legal side, there is no specific information in the laws on how to act in such a case. The processing of personal data, medical analyses and scientific research – all this goes against the definition of AI and the algorithm of its work [11].
There are several international organizations that develop rules and regulations for the use of artificial intelligence in medicine and science, but this is still not enough. There must be specialists and competent people in the field of regulation and control of AI actions. One of the modern such organizations is the World Health Organization (WHO). In October 2023, WHO presented the principles of regulation of AI technologies in healthcare [15]. The document indicates the need to conduct an examination of the safety and effectiveness of AI systems and to ensure a dialogue between stakeholders - developers of such systems, regulatory authorities, manufacturers, doctors and patients. WHO believes that AI technologies can be useful in conditions of shortage of medical personnel, and they are also able to facilitate, accelerate and optimize work based on the cooperation of the AI system and specialists. "At the same time, the process of mastering artificial intelligence technologies, including large language models, is happening rapidly, sometimes without a thorough understanding of the mechanisms of their operation, which can bring both benefits and harm to end users, including specialists and patients. When working with medical data, AI systems can use personal information, the confidentiality and integrity of which must be protected using reliable legal and regulatory mechanisms," the WHO believes [15]. 
The use of AI also affects social aspects. Will AI be able to replace the work of doctors and scientists? Will the progress of AI affect the reduction of people's jobs? At the moment, AI has the potential to automate and optimize many tasks, including the work of doctors and scientists. However, AI currently cannot completely replace humans in these professions. A doctor or scientist needs not only practical skills, but also the ability to make difficult decisions, analyze the context and use empathy towards colleagues and patients. AI can be a useful tool to support doctors and scientists by helping them make decisions and reducing the time required to complete certain tasks.
Regarding the impact of AI progress on the labor market, there are concerns that automation may lead to job cuts for people. However, historical experience shows that at the same time as new technologies destroy certain jobs, they also create new opportunities and new professions. Instead of replacing and reducing, AI can help people focus on higher-level tasks that require creative thinking and interpersonal skills.
In any case, the introduction of AI into work processes must be carried out taking into account the social and economic consequences in order to ensure fairness, accessibility and equality for all citizens.

Prospects for the use of AI and new opportunities
When discussing the use of artificial intelligence in medicine and science, the following areas of change in the long term can be identified:
1. Legal aspect and regulation: 
The development of AI in medicine and science poses new questions and challenges in the field of law and ethics. It is necessary to update the legislative framework in order to regulate the area of responsibility of AI, protect data privacy and ensure ethical principles for the use of AI. External data validation and a clear understanding of the intended use of AI are also important. The new regulations should ensure safety and simplify regulation, as well as address the issue of area of responsibility, patient confidentiality, protection of research data, as well as promote health insurance and assistance to patients affected by AI actions. Regular data quality control plays an equally important role: for example, it is important to ensure that systems do not reinforce biases and errors in diagnosis, analysis and recommendations. To manage risks, it is necessary to comprehensively address issues such as "intended use", "continuous learning", intervention by specialists, training models, new methodological developments on working with AI systems and cybersecurity threats.
Open questions: 
1. How to ensure a balance between autonomous AI solutions and the role of doctors in decision-making?
2. What rules should exist for the proper use and transparency of AI research in medicine and science?
3. What encryption methods are used to protect medical data when using AI?
4. How can doctors be trained to use AI systems in medical practice?
2. Ethical and social aspects: 
The use of AI may raise ethical issues regarding the protection of patient privacy, accessibility and fairness of the use of AI in medicine and science. The answers to these questions require fundamental discussion and the development of appropriate norms and standards.
Open questions: 
1. What measures are being taken to protect patient data when using AI in medicine?
2. What aspects of ethics relate to the use of AI in medicine, and what measures can be taken to ensure the ethical use of AI in this area?
3. How can organizations cope with the problems associated with changes in legislation to regulate the use of AI in medicine and science in Russia, and who will be responsible?
4. What technological innovations are needed to fully realize the potential of AI in medicine and science?
5. How will AI progress affect medical personnel in the future?
3. Scientific perspectives: 
The use of AI can open up new scientific perspectives, as it can process and analyze huge amounts of data and identify patterns that are not accessible to humans. AI can help in the discovery of new knowledge, hypotheses, and facilitate research work. AI can speed up the process of developing new drugs. Machine learning algorithms can analyze large amounts of data on molecular structure and pharmacological properties to identify potentially effective compounds, reducing research costs and time.
4. Medical perspectives: 
The use of AI in medicine opens up prospects for early diagnosis of diseases, effective treatment and improvement of the quality of medical care. AI can help doctors make decisions, analyze medical images and patient data, and optimize treatment strategies. AI also makes it possible to develop personalized treatment approaches. By analyzing patient data, including genetic information, medical history, and test results, AI can predict which type of treatment will be most effective for a particular patient. The main function of AI is a tool for the support and assistance of doctors and specialists. AI systems can help analyze medical images, make preliminary diagnoses, offer treatment recommendations and warn about possible complications, which can improve the quality and effectiveness of medical care. AI can provide guidance and decision support in the treatment of patients. By analyzing patient data, research, and treatment protocols, AI can help a doctor choose the best medical solutions and provide an individual approach to each patient.
5. Technological changes: 
The application of AI requires the development and improvement of technologies such as machine learning algorithms, cloud computing, large amounts of data and computing power. Technical innovations and improvements will unlock the full potential of AI in medicine and science. AI can automate routine tasks and processes in scientific research. This can include analyzing large amounts of data, processing information, creating models and formulating hypotheses, which allows researchers to more freely engage in semantic data analysis and generate new ideas.
These are just some possible options for the structure of changes that contribute to the optimization of AI in medicine and science, the removal of white spots in legislation and regulation of this area. They emphasize the importance of taking into account aspects of a legal, ethical, social, scientific and technological nature when developing strategies for using AI to maximize efficiency and solve urgent problems in these areas.

 
Conclusion
The use of AI in medicine and science offers a wide range of perspectives and new opportunities, including early diagnosis, personalized treatment, improving the quality of medical care and automating research processes. These applications of AI can significantly improve the efficiency, accuracy of results and speed of work in the field of medicine and science.
    The development of AI in medicine and science also raises ethical and social responsibility issues, such as the privacy of patient data, the use of AI in ethical decision-making, and the trust of doctors, scientists, and patients in automated systems. It is necessary to develop appropriate regulatory and legal frameworks to regulate the use of AI in these areas. 
The development and implementation of AI requires consideration and discussion of various aspects in order to ensure its effective and responsible use.