Knowledge of how Artificial Intelligence (AI) operates and how it can be utilized is gradually ceasing to be a distinguishing competitive advantage and is becoming one of the fundamental requirements for functioning in today’s reality. The KPMG report „Artificial Intelligence in Poland” indicates that 69% of respondents in Poland use AI regularly, with 54% of working individuals utilizing AI for professional purposes. Given the growing interest in this technology and the continuous emergence of new methods for its exploitation, it is essential to remain up to date with current knowledge.

This subsequent publication in a series dedicated to artificial intelligence focuses on areas of daily life that have undergone significant changes in recent years due to the popularization and widespread availability of AI systems. Specifically, it analyzes changes in the labor market, the creative sector, and the defense industry. Specialized applications of AI in law and medicine are also discussed.

Labor Market: Recruitment and Process Management in Enterprises

The labor market is an area undergoing dynamic transformation as a result of integrating AI into professional processes. AI tools are utilized by HR teams in recruitment processes, including the creation of job descriptions, preliminary screening of CVs, and communication with candidates. As highlighted by Hays, a leader in specialist recruitment and HR solutions, advanced AI systems can streamline and shorten the duration of the recruitment process.

However, it must be borne in mind that the AI Act classifies AI systems used in the areas of employment, workers management, and access to self-employment as high-risk AI systems. Following the entry into force of the Act’s regulations, this will entail strict limitations regarding the methods of utilization and management of such tools. Business operators using these systems should implement the procedures stipulated in the Act as soon as possible (which concern, among others, human oversight, logging/monitoring of events, or conducting data protection impact assessments in compliance with GDPR) to avoid severe administrative fines. In practice, the most crucial aspect for a recruiter will be ensuring that the AI system serves exclusively as a supportive tool and does not replace human judgment in the candidate selection process.

AI systems are deployed in the labor market beyond recruitment alone. Virtual assistants, namely AI systems designed to interact with clients, can respond to simple inquiries and redirect more complex matters to employees. Consequently, specialists can focus exclusively on intricate issues, leaving basic questions to AI systems. A study by the National Bureau of Economic Research (NBER) conducted on customer service consultants revealed that implementing AI systems in the form of a virtual assistant increased productivity by an average of 14%. Notably, researchers also recorded an improvement in the quality of interactions with clients.

Under the AI Act, however, it must be remembered that users must always be aware of when they are interacting with customer service and when they are communicating with a virtual assistant. It is therefore mandatory for virtual assistants to inform users, prior to commencing an interaction, that the conversation is not being conducted with a human being.

According to a report by the NASK National Research Institute, over 5 million jobs could be partially automated using AI systems. These statistics may cause concern among workers; the report indicates that over 800,000 individuals work in occupations with the highest exposure to the impact of AI. Nevertheless, the authors of the study believe that the increase in productivity and creativity brought about by the application of AI systems represents an opportunity to build a more equitable labor market.

AI and Creativity: Easier Access to to Content, Risks to Copyright Law

Artificial intelligence has abruptly impacted activities tied to creative work. As recently as 2022, most large language models could only generate text; however, contemporary models can generate audio, images, video materials, as well as audiovisual translations.

This has a particular impact on the marketing industry. According to a HubSpot report, a company specializing in the development and sale of CRM (Customer Relationship Management) software, 66% of marketers use AI in their work, and 40% utilize image generators. AI also plays a role in content distribution and optimization. A case in point is the initiatives undertaken by the American company Meta. AI systems implemented within social media platforms such as Instagram and Facebook have been used for several years to predict user interests based on their interactions, such as searched pages, likes, or shares. These systems find commercial application, as the interactions also influence the nature of advertisements displayed to the user.

In 2025, Meta also introduced the „Meta AI Translation” tool into widespread use. This tool is capable of autonomously translating short videos shared by an author into audio format, utilizing an imitation of their voice. The language is adapted to the recipient’s language; thus, a short video recorded in English can be displayed to a viewer in their native language. The process itself requires no input from the creator, and this solution significantly expands the potential audience. According to some researchers, AI could automate up to 26% of tasks related to art, design, media, and the entertainment industry.

AI also influences artistic creation. Growing attention is being drawn to musical tracks generated entirely by artificial intelligence. An example is the musical track „Walk My Walk,” released by „Breaking Rust”, a project created using artificial intelligence, which reached number one on the Billboard Country Digital Song Sales chart. This track accumulated over 3 million streams in less than a month.

Creativity generated by artificial intelligence remains difficult to reconcile with classical constructs of copyright law. It remains unclear whether AI models retain fragments of training data and whether they can reproduce content similar or identical to protected works. Another remaining problem is the issue of authorship of works generated by AI; Polish jurisprudence maintains the position that an author can only be a natural person. This means that authorship, and consequently economic copyright, cannot be attributed to an AI system. Whether a creation generated by AI can belong to the individual utilizing a given system has not been unequivocally resolved. Eventual authorship will depend on the nature and scale of the human’s creative contribution to the artificially generated content.

AI in Medicine: Supporting Diagnostics

Healthcare is one of the sectors where the application of artificial intelligence holds particularly high potential. AI is utilized, among other areas, in diagnostic imaging, patient data analysis, and supporting the work of medical personnel. One documented example involves research into the use of AI in mammography. According to the MASAI study conducted by Lund University, screening examinations supported by artificial intelligence systems allowed for the detection of 29% more breast cancer cases compared to traditional screening methods. Assisted screening additionally reduced the workload of radiologists without increasing the rate of false-positive results.

Applications of AI in healthcare require particular caution, however. A study published in Science revealed that a widely used healthcare algorithm affecting millions of patients systematically underestimated the health needs of Black patients. After correction, the percentage of individuals qualified for additional care rose from 17.7% to 46.5% (Obermeyer et al., 2019). A University of Oxford study published in Nature Medicine also indicates that artificial intelligence poses a risk to individuals in need of a medical opinion due to the tendency of AI to present inaccurate and inconsistent information. Therefore, not only is oversight by medical personnel required, but it is also necessary to inform individuals seeking medical care about the risks stemming from diagnoses generated by AI systems.

AI in Law: Assistance for the Lawyer, Not a Substitute for Professional Judgment

Artificial intelligence is also transforming the work of legal professionals. AI tools are used for document analysis, case law research, drafting initial versions of pleadings, summarizing case files, comparing contracts, creating checklists, and supporting due diligence processes. According to the Future Ready Lawyer 2024 report by Wolters Kluwer, 76% of corporate lawyers and 68% of law firm lawyers use generative AI at least once a week.

Additionally, as indicated by the report of the Working Group on AI, artificial intelligence is utilized in the judiciaries of certain countries: in Brazil, the VICTOR system assists the Supreme Court in conducting admissibility tests to evaluate motions submitted to the court, while in England and Wales, the VIQ system records audio, converts speech to text, and transcribes hearings in county courts and judgments in common courts.

This does not mean, however, that AI can replace the legal analysis performed by professionals. Research by Stanford HAI indicates that hallucinations in legal tasks are pervasive. In tests concerning specific legal questions, general models exhibited very high error rates, ranging from 58% to 88% depending on the model and task. Newer analyses of specialized legal AI tools also indicate a significant percentage of erroneous answers: over 17% in the case of Lexis+ AI and Ask Practical Law, and over 34% in the case of Westlaw AI-Assisted Research. Added to this are issues of professional secrecy, confidentiality of client data, personal data protection, professional liability, and transparency toward the client. In the context of the justice system, the AI Act classifies certain systems used by judicial authorities to examine, interpret, and apply the law to the facts of a case as high-risk systems.

AI and the Defense Sector: Security and the Boundaries of Autonomy

The defense sector is a distinct area of AI application; AI systems are utilized, among others, in the analysis of intelligence, surveillance, and reconnaissance data, referred to as ISR. Their task can involve organizing large datasets, detecting anomalies, classifying objects, and supporting commanders in building situational awareness.

The Polish Armed Forces utilize unmanned systems and are developing reconnaissance-strike systems. A separate area of AI application in defense is cybersecurity. AI systems can support network traffic analysis, threat classification, and accelerate incident response.

The most severe risks concern applications related to the use of force. The Stockholm International Peace Research Institute (SIPRI) indicates that autonomous weapon systems and AI-enabled decision support systems influence the human role in the military targeting process. The greater the influence of an AI system on decision-making, the more significant becomes the ability to explain the system’s operation, document the decision-making process, and assign responsibility for the consequences of its use.

Summary:

The proliferation and rising utility of artificial intelligence have led to the presence of AI systems across most areas of social and professional life. AI tools influence work organization, content distribution, assist doctors and lawyers in their daily duties, and strengthen state security. In each of these areas, the potential of AI is significant.

Concurrently, the very attributes that render AI useful create parallel legal, organizational, and social risks. The key challenge is not the deployment of AI systems itself, but ensuring that these systems are safe, controlled, and legally compliant. This requires appropriate legal frameworks, risk assessment procedures, data quality control, and human oversight. It is also essential to develop widespread knowledge of AI operations and applications, as this technology will continue to gain significance in the future.

Sources and reference material:

• KPMG Report: https://assets.kpmg.com/content/dam/kpmg/pl/pdf/2025/07/pl-Raport-KPMG-w-Polsce-KPMG-AI-Trust-2025-web.pdf
• Hays Article: https://www.hays.pl/blog/insights/jak-wykorzystuje-sie-sztuczna-inteligencje-w-rekrutacji-
• AI Act Regulation: https://eur-lex.europa.eu/legal-content/PL/TXT/PDF/?uri=OJ:L_202401689
• NBER Study: https://www.nber.org/papers/w31161
• NASK Study: https://www.nask.pl/aktualnosci/genai-zmienia-rynek-pracy-raport-nask-i-ilo-o-przyszlosci-zatrudnienia
• HubSpot Report: https://blog.hubspot.com/marketing/state-of-ai-report…
• Meta AI: https://about.fb.com/news/2025/10/discover-reels-around-world-meta-ai-translation/?utm_source=chatgpt.com
• Universitat Oberta de Catalunya Study: https://www.uoc.edu/en/news/2025/ai-could-automate-creative-professions
• Billboard Country Digital Song Sales
• AAAI Study: https://ojs.aaai.org/index.php/AIES/article/view/31697
• Lund University Study: https://www.lunduniversity.lu.se/article/ai-supported-breast-cancer-screening-new-results-suggest-even-higher-accuracy?utm_source=chatgpt.com
• GOV AI Study: https://ai.gov.pl/ai-dla-ciebie/artykuly/jak-sztuczna-inteligencja-zmienia-medycynę…
• JMIRS Study: https://www.jmirs.org/article/S1939-8654(23)00079-6/fulltext
• Annals of Family Medicine – Article on AI Hallucinations: https://www.annfammed.org/content/hidden-risk-ai-hallucinations-medical-practice
• Future Ready Lawyer Report Wolters Kluwer: https://www.wolterskluwer.com/pl-pl/know/future-ready-lawyer-2024
• AI Working Group Report: GRAI
• Stanford HAI Studies: https://hai.stanford.edu/news/hallucinating-law-legal-mistakes-large-language-models-are-pervasive?utm_source=chatgpt.com
• Polish Armed Forces Source: https://www.gov.pl/web/obrona-narodowa/modernizacja-techniczna-szrp
• SIPRI Publication: https://www.sipri.org/publications/2025/other-publications/autonomous-weapon-systems-and-ai-enabled-decision-support-systems-military-targeting-comparison-and