E-ISSN: x-x
ARTICLE
Business process transformation to maintain high performance:
Driving an AI-based business value project Elfan Kaukab 1.
Ali Akbar Anggara 2 *.
Ali Imron 3 Faculty of Economics and Business.
Universitas Sains Al-QurAoan.
Wonosobo.
Indonesia.
elvankaukab@yahoo.
Faculty of Economics and Business.
Universitas Muhammadiyah Purwokerto.
Purwokerto.
Indonesia.
aliakbarang@ump.
Faculty of Creative Design and Digital Business.
Institut Teknologi dan Sains Nahdlatul Ulama Pekalongan.
Pekalongan.
Indonesia.
alialta@gmail.
Correspondence: aliakbarang@ump.
Abstract This study investigates the impact of Artificial Intelligence (AI) on firm performance, emphasizing the business value generated by AI-enabled transformation projects.
Employing a four-step sequential approach, including .
analysis of AI concepts and technologies, .
in-depth review of cross-sector case studies, .
data collection from AI solution providersAo databases, and .
literature review.
This research draws on the theory of IT capabilities to examine AIAos influence at organizational and process levels.
The analysis is based on 500 case studies from sources including IBM.
AWS.
Cloudera.
Nvidia.
Conversica, and Universal Robots.
Findings reveal that AI, encompassing technologies such as machine translation, chatbots, and self-learning algorithms, enhances business performance by optimizing processes, automating operations, improving information flows, and enabling predictive and interactive However, performance gains are realized only when organizations leverage AI features to reconfigure and innovate their processes.
AI adoption thus emerges not merely as a technological upgrade but as a driver of strategic transformation and competitive advantage.
The study offers both theoretical and managerial contributions.
Theoretically, it proposes a model for assessing AIAos business value, addressing gaps in the literature.
Managerially, it guides decision-makers in aligning data, talent, domain expertise, partnerships, and scalable infrastructure to maximize AI benefits.
By viewing AI as an integrated set of IT configurations rather than a standalone tool, organizations can achieve higher performance, enhance investment returns, and strengthen competitive positioning.
These insights position AI as a critical enabler for new business models and sustainable organizational growth.
Keywords: artificial intelligence, business process, business value.
IT capabilities, firm performance Academic Editor: Gara Huang Received: August 01, 2025 | Revised: Agust 07, 2025 | Accepted: August 11, 2025 | Published: August 12, 2025 Citation: Kaukab.
Anggara.
, & Imron.
Business process transformation to maintain high performance: Driving an AI-based business value Current Perspective on Business Operations, 1.
, 1-20.
Introduction Information technologies (IT.
are now a fundamental component of modern professional practices, reshaping and, in many cases, disrupting essential processes and Current Perspective on Business Operations 2025, 1.
, 1-20.
Current Perspective on Business Operation 2025, 1.
, 1-20.
operations across sectors (Devaraj & Kohli, 2003.
Nwamen, 2.
When embedded within a business ecosystem.
IT influences how organizations engage with customers, potential clients, and strategic partners (Lauterbach, 2019.
Nwamen, 2.
, while also driving the evolution of internal workflows and operational structures.
Within this technological landscape.
Artificial Intelligence (AI) has emerged as a particularly transformative force, achieving rapid progress unmatched by most other innovations in recent decades (Blanchet, 2016.
Lee et al.
, 2018.
Wiljer & Hakim, 2.
AI is broadly described as a suite of concepts, methods, and tools designed to create systems capable of emulating intelligent behavior with limited human input (Benko & Lynyi, 2009.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
Industry projections reflect AIAos accelerating importance in business transformation.
By 2019.
AI technologies were integrated into approximately 40% of digital transformation projects, and it was anticipated that 75% of enterprise applications would incorporate AI functionalities by 2021 (Crews, 2.
Organizations are turning to AI to enhance efficiency, stimulate innovation, and develop new services (CIGREF, 2016, 2018.
Crews, 2.
Global technology leaders such as the U.
-based GAFAM and ChinaAos BATX have committed substantial resources to advanced AI domains, particularly machine learning and deep learning (Lee et al.
, 2018.
PwC, 2019.
Vochozka et al.
, 2.
The explosive growth of AI, propelled by the digitization of industries and the widespread generation of data from connected devices (Dopico et al.
, 2016.
Sheth, 2.
, underscores its potential as a driver of future competitive advantage (PwC, 2018, 2019.
Tractica, 2.
Although AI does not replicate the full complexity of human cognition, it demonstrates exceptional performance in specialized tasks, yielding significant organizational benefits (Blanchet, 2016.
Lee et al.
, 2018.
Wiljer & Hakim, 2.
This research seeks to address a critical gap by evaluating how AI capabilities contribute to improvements in both organizational and process-level performance.
Drawing on prior academic work and an analysis of 500 case studies from AI vendors and industry publications, the study assesses the business value of AI-enabled transformation initiatives.
The paper proceeds by outlining the material and method (Section .
, findings (Section .
, discussion (Section .
, and final conclusions, study limitations and implications (Section .
Materials and Methods Evolution of AI The origins of Artificial Intelligence (AI) trace back to the early days of computer science when pioneers such as Alan Turing explored the concept of machine intelligence.
His influential works including Intelligent Machinery .
and Computing Machinery and Intelligence .
, laid the groundwork for decades of subsequent AI research (Turing, 2.
AI, in its broadest sense, refers to the deployment of technological systems designed to replicate human cognitive functions in order to achieve specific goals autonomously, while adapting to contextual constraints (Benko & Lynyi, 2009.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
Over time, multiple landmark studies and innovations have shaped the evolution of AI, as summarized in Table 1.
Table 1.
Artificial Intelligence Evolution Date AuthorsAo contribution to the development of AI 1940Ae1956 Norbert Wiener pioneered the field of cybernetics, exploring how the human mind functions and proposing it could be represented as a Aublack boxAy in which behavior is regulated by feedback loops.
His work suggested that the brainAos billions of cells could be expressed through mathematical models.
Current Perspective on Business Operation 2025, 1.
, 1-20.
Date AuthorsAo contribution to the development of AI an idea later advanced by Donald Hebb, who contributed to the development of formal neurons with adaptive learning abilities (Brown & Milner, 2.
In 1943.
Warren McCulloch and Walter Pitts introduced the first mathematical representation of a biological neuron, applying a physiological framework to AI research (Benko & Lynyi, 2009.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
Herbert Simon, in 1947, put forward the concept of Aubounded rationality,Ay while in 1945.
Allen Newell explored heuristic problem-solving, an empirical approach whose validity is context-dependent.
Their combined contributions highlighted the reciprocal influence between AI and computer science: computing advancements enabled AI experiments, while AI challenges stimulated the creation of new computing tools (Benko & Lynyi, 2009.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
Between 1937 and 1948.
Claude Shannon linked Boolean algebra with electrical circuitry, laying the foundation for digital electronics and information theory (Verdu, 1.
Nathaniel Rochester developed the first symbolic assembler language and was instrumental in AI projects such as LISP, the Geometry Theorem Prover, and symbolic assembly (Gelernter et al.
, 1958.
Rochester et al.
, 1.
In 1949.
Hebb formulated the learning rule for artificial neurons, enabling memory functions to operate as feedback processes in neural networks and forging a connection between cognition and language (Benko & Lynyi, 2009.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
John McCarthy applied logic to the creation of thinking machines (Kline, 2010.
McCarthy, 1.
, while Marvin Minsky proposed a schematic model for artificial neural networks (Minsky, 2007.
Minsky & Papert, 1.
By 1954, researchers at Georgetown University produced a language translation program capable of processing several dozen basic sentences using a 250-word vocabulary and six grammar rules, executed on an IBM 701 (Hutchins, 2.
Contributions from John von Neumann on computer architecture and Alan Turing on computable functions further advanced AI theory (Benko & Lynyi, 2009.
Godfrey & Hendry, 1993.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
In 1956.
Newell.
Simon, and Shaw developed the Information Processing Language (IPL), which used list structures to simulate associative memory (Benko & Lynyi, 2009.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
Other notable advancements included Ray SolomonoffAos work in machine learning and algorithmic probability (Solomonoff, 1.
and Oliver SelfridgeAos early research in expert systems and pattern recognition (Husbands et al.
, 2.
Although AI ideas had appeared in popular culture for years, the fieldAos formal inception is widely attributed to the 1956 Dartmouth Conference led by McCarthy (McCarthy et al.
, 2.
This period marked the beginning of AIAos Augolden age,Ay characterized by strong public investment in research (Buchanan, 2.
Development of MYCIN, the first expert system for diagnosing bacterial infections and recommending antibiotics (Shortliffe et al.
, 1.
1974Ae1980 During this period, the field of AI faced what became known as its first Auwinter.
Ay Ambitious goals collided with insufficient resources, leading to stalled progress and reduced enthusiasm in research (Benko & Lynyi, 2009.
Godfrey & Hendry, 1993.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
1980Ae1987 AI research regained momentum with the emergence of more capable intelligent systems.
Notably.
Kai-Fu Lee and Sanjoy Mahajan created BILL, a Bayesian learning program designed to play the board game Othello, reflecting the eraAos renewed optimism and technical progress (Benko & Lynyi, 2009.
Godfrey & Hendry, 1993.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1.
Current Perspective on Business Operation 2025, 1.
, 1-20.
Date AuthorsAo contribution to the development of AI 1987Ae1993 The collapse of the specialized hardware industry triggered AIAos second Auwinter,Ay characterized by funding cutbacks and slowed development (Benko & Lynyi, 2009.
Haenlein & Kaplan, 2019.
McCorduck et al.
, 1977.
Wamba et al.
, 2.
1993Ae2011 AI shifted towards a data-driven paradigm, fueled by growing computing capabilities (Buchanan.
A landmark event occurred in 1997 when IBMAos DeepBlue chess system defeated world champion Garry Kasparov (Campbell et al.
, 2.
The combination of greater data availability, improved connectivity, and more powerful devices accelerated advancements, contributing to a surge in AI-related patent filings after 2012 (Shoham, 2.
AI and Business Since 2010.
AI has entered a renewed phase of rapid growth, driven by advancements in computing power and unprecedented access to vast datasets (CIGREF, 2018.
PwC.
This resurgence can be attributed to three key developments: .
the creation of highly sophisticated algorithmic models, .
the widespread availability of affordable graphics processing units capable of executing massive computations within milliseconds, and .
the emergence of extensive, accurately annotated datasets enabling advanced learning in intelligent systems (Jain et al.
, 2004.
Khashman, 2009.
PwC, 2.
Core AI technologies including machine learning, deep learning, chatbots, neural networks, and virtual assistants are reshaping corporate processes and organizational structures (CIGREF, 2018.
Kuzey et al.
, 2014.
PwC, 2.
They have altered how firms manage information, interact with their environment, and deliver value.
However, capitalizing on these opportunities requires not only technological investment but also cultural, strategic, and skills-based transformation (Di Francescomarino & Maggi, 2020.
Lee et al.
, 2018.
Sikdar, 2.
An example is IBMAos Watson platform, which can process natural language queries to support diverse domains such as marketing, management, legal services, and healthcare (Kohn et al.
, 2.
Watson Health, in particular, enables physicians to integrate patient records with up-to-date medical data, enhancing diagnostic precision and care personalization (Kohn et al.
, 2.
AIAos conceptual foundations date back to the 1960s, but its major breakthroughs have emerged in the 2000s with the rise of machine learning and, more recently, deep learning, which uses multi-layered neural networks (Buchanan, 2005.
PwC, 2019.
Yoav Shoham.
These models learn patterns by processing millions of examples, ranging from medical diagnoses and energy reserve predictions to pricing forecasts (PwC, 2019.
Zemouri et , 2.
When combined with big data, such systems can perform complex operations faster and more accurately than humans.
The applications are broad: autonomous vehicles in transport (Falcone et al.
, 2.
disease detection in healthcare (Jiang et al.
, 2017.
Koh & Tan, 2.
, conversational agents for customer service (Rubin et al.
, 2.
, natural language processing for email automation (Gabrilovich & Markovitch, 2.
, facial recognition in security, and data-driven urban planning for smart cities (Srivastava et al.
, 2.
AIAos integration spans the entire value chain from R&D, maintenance, operations, and logistics to sales, marketing, demand forecasting, and after-sales services (Kuzey et al.
, 2014.
PwC, 2.
As a strategic growth driver.
AI offers organizations several advantages:
Enhancing operational efficiency, optimizing supply chains, improving customer experience, and refining product/service features, including recommendation systems (Kuzey et al.
, 2014.
PwC, 2.
Current Perspective on Business Operation 2025, 1.
, 1-20.
Facilitating rapid adaptation to market changes, enabling the creation of new business models, and improving forecasting and planning capabilities (Kuzey et al.
PwC, 2.
Detecting fraud, automating cybersecurity threat analysis, streamlining IT processes, and optimizing sales workflows (CIGREF, 2018.
PwC, 2.
Supporting medical diagnostics and treatment planning, predicting disease progression, recommending personalized therapies, and aiding in epidemic prevention and pharmaceutical safety (Jiang et al.
, 2017.
Johnson et al.
, 2.
Automating quality control, asset management, and logistics across industries (Di Francescomarino & Maggi, 2020.
Rubin et al.
, 2010.
Sikdar, 2.
Market forecasts reinforce AIAos economic significance.
According to Tractica .
, global AI revenues are projected to approach $90 billion by 2025, up from $7 billion in 2018, with North America.
Europe, and AsiaAePacific as primary beneficiaries.
Three areas are expected to dominate revenue generation: .
moving object detection and avoidance, .
static image recognition and classification, and .
medical data processing.
Together, these domains could generate nearly C21 billion in revenue between 2016 and 2025.
IT Capabilities In an increasingly globalized and competitive market, innovation, quality, and rising customer expectations have pushed organizations to embed information technology (IT) into their management strategies (Bolwijn et al.
, 2018.
Rachinger et al.
, 2.
Historically, automation began with assembly lines replacing manual labor.
today, technological advances enable machines to handle more complex, value-adding tasks (Dopico et al.
, 2016.
Lee et al.
, 2.
Modern organizations face the challenge of innovating in rapidly changing markets where information is a critical strategic asset (Kuusisto, 2.
IT reshapes business processes, demanding changes in culture, mindset, and skills to fully leverage its potential (Devaraj & Kohli, 2003.
Turulja & Bajgoric, 2.
AI, as part of the IT ecosystem, enhances performance, strengthens customer and partner relationships, and optimizes operations (Kelly et al.
, 2.
The Aueconomy of powerAy paradigm suggests that market positioning depends on resources that both secure and enhance competitive advantage (Liu et al.
, 2.
This raises two central questions: how IT capabilities improve organizational performance, and how they generate business value in transformation projects (Abijith & Wamba.
Research since the 1980s shows mixed evidence, some finding no significant effect (Koski, 1.
, others reporting positive impacts (Devaraj & Kohli, 2.
Six main approaches link IT to performance:
A Economic theory Ae measuring ITAos economic contribution (Lehr & Lichtenberg, 1.
A Social psychology Ae assessing effects on individuals and groups (Davis, 1.
A Competitive analysis Ae evaluating industry and market impacts.
A Strategic alignment Ae aligning IT with organizational strategy (Henderson & Venkatraman, 1.
A Process-based Ae viewing IT as value-creation processes.
A Resource-based Ae leveraging unique, valuable resources for competitive advantage (Barney et al.
, 2.
From a resource-based view.
IT capabilities combine tangible assets, intangible assets, and human expertise to create value.
These can be assessed through three dimensions: .
IT management capabilities, .
IT staff expertise, and .
IT infrastructure flexibility (Abijith & Wamba, 2012.
Kim et al.
, 2.
Hypothesis Development Our research model was developed in two stages: first, reviewing literature on ITAos role in enhancing performance at both organizational and process levels .
Abijith & Current Perspective on Business Operation 2025, 1.
, 1-20.
Wamba, 2012.
Kim et al.
, 2011.
Mooney et al.
, 1.
, and second, analyzing mini-case studies to assess the business value of AI-enabled transformation.
Six hypotheses emerged, linking AI capabilities to performance improvement (Figure .
Figure 1.
Proposed Research Model AI capabilities refer to an organizationAos capacity to combine its organizational, human, and AI-specific resources to generate and capture business value (Abijith & Wamba, 2012.
Kim et al.
, 2011.
Liu et al.
, 2.
Resources within a firm may be organizational, human, material, or intangible (Kim et al.
, 2011.
Liu et al.
, 2.
Drawing from prior studies, this research focuses on three main AI resource dimensions.
AI Management Capability (AIMC) is the organizations and its employeesAo ability to oversee and model intelligent behaviors in technological systems to enhance long-term value and sustainability.
This capability supports strategic planning, strengthens intra- and inter-organizational relationships, guides investment decisions, and facilitates coordination and control (Ha & Jeong, 2010.
Hamet & Tremblay, 2017.
Kim et al.
, 2.
Kim et al.
further emphasizes that effective control, driven by staff expertise can improve infrastructure flexibility.
Hyphotesis 1: AI management capability contributes to enhance overall AI capabilities, which in turn enhance AIAos impact at the process level.
Current Perspective on Business Operation 2025, 1.
, 1-20.
AI Personal Expertise (AIPE) refers to the specialized skills and knowledge of organizational staff in AI technologies, business operations, and interpersonal competencies required to model or apply intelligent behavior in technological systems (Ha & Jeong.
Hamet & Tremblay, 2017.
Jiang et al.
, 2017.
Kim et al.
, 2.
For IT teams, a balanced mix of technical awareness, ownership, integration, and management abilities which combined with knowledge of core IT components.
It is essential for optimizing AI resource utilization.
The ability to align AI initiatives with organizational strategy is a key driver of business value creation, and this alignment strengthens when staff possess the right skill set.
When employees understand how strategic goals integrate with AI and IT capabilities, their expertise becomes a valuable intangible asset (Abijith & Wamba.
Kim et al.
, 2011.
Liu et al.
, 2.
Organizations with highly skilled AI professionals are better positioned to adapt to dynamic environments, align AI with strategic objectives, and develop efficient, reliable intelligent systems.
Hyphotesis 2: AI personal expertise contributes to enhance AI capabilities, which in turn enhance AIAos impact at the process level.
AI Infrastructure Flexibility (AIIF) encompasses the full range of technological resources including software, hardware, data, systems, networks, and telecommunications which needed to operate AI solutions effectively (Kim et al.
, 2011.
Liu et al.
, 2013.
Wamba et al.
, 2.
Flexible infrastructure enables organizations to quickly adjust system components and adapt to shifts in market conditions, strategic partnerships, acquisitions, or global collaborations (Abijith & Wamba, 2012.
Bhatt et al.
, 2010.
Fink & Neumann, 2.
Successful AI deployment depends on aligning several critical factors: high-quality data, combined ITAeAI talent, domain expertise, effective decision-making, external collaborations, and scalable infrastructure.
While the first four elements serve as the Aufuel,Ay scalable infrastructure acts as the AuengineAy that powers AI operations.
Robust and adaptable infrastructure allows efficient use of IT resources and supports organizational restructuring when integrating AI technologies.
Advanced capabilities such as self-configuration, self-healing, and self-optimization which help prevent operational issues, drive innovation in business processes, and improve resource utilization (Kim et al.
, 2011.
Liu et al.
Wamba et al.
, 2.
Hyphotesis 3: AI infrastructure flexibility contributes to enhance AI capabilities, which in turn enhance AIAos impact at the process level.
The automation effect refers to AIAos ability to replace or significantly reduce manual processes through standardized, repeatable instructions.
This shift increases operational efficiency, enhances process reliability, and reduces labor costs by automating previously paper-based or human-intensive tasks (Abijith & Wamba, 2012.
Kim et al.
, 2011.
Liu et , 2.
Hyphotesis 4: The automation effect contributes at the process level, which is positively associated with the influence of AI at the organizational level.
The informational effect describes AIAos capacity to gather, store, process, and share data within and across organizations.
By converting high-quality data into timely, accurate, and relevant insights.
AI enables faster and better decision-making, strengthens resource control, enhances coordination, and improves responsiveness among staff.
These capabilities contribute to both financial stability and operational efficiency (Abijith & Wamba.
Kim et al.
, 2011.
Liu et al.
, 2013.
Mooney et al.
, 1.
Hyphotesis 5: The informational effect contributes at at the process level, which is positively associated with the influence of AI at the organizational level.
Current Perspective on Business Operation 2025, 1.
, 1-20.
The transformational effect reflects AIAos role in driving innovation and reengineering organizational processes.
By enabling service enhancement, redesigning workflows, and building competitive capabilities.
AI fosters strategic change and strengthens customer These innovations help organizations adapt to dynamic markets and create new products or services, positioning AI as a catalyst for long-term business transformation (Abijith & Wamba, 2012.
Kim et al.
, 2011.
Mooney et al.
, 1.
Hyphotesis 6: The transformational effect contributes at the process level, which is positively associated with the influence of AI at the organizational level.
Methods This study followed a three-phase process, as illustrated in Figure 2.
The first phase was conceptual, focusing on selecting the target case studies for analysis.
The second, a refinement and development stage, involved validating the accuracy and reliability of the case study data.
The final assessment phase evaluated these cases to extract both practical insights and theoretical contributions to the IT capability framework.
An archival data analysis approach was employed, leveraging secondary data from documented case studies.
In the current digital era, such data are widely accessible, costefficient, and suitable for in-depth analysis.
This method is particularly appealing given that valuable, high-cost primary data often remain underutilized.
Similar approaches have been used in prior research, such as Abijith and Wamba .
in examining the AuBusiness Value of RFID-Enabled Healthcare Transformation Projects,Ay and Faure et al.
, who analyzed 13 case studies to explore the impact of various agricultural research models in developing countries.
Figure 2.
Research Model Development Results An in-depth review of the case studies confirms that artificial intelligence (AI) and its related technologies provide diverse opportunities, advantages, and services aimed at enhancing organizational performance.
As outlined in Table 2.
AI adoption delivers value across multiple business areas by transforming processes into intelligent, optimized, selfadjusting, and automated systems, replacing many resource-intensive, paper-based, and manual operations.
Table 2.
AI-value Based Transformation Project Industry Common Benefits of AI-Enabled Organizational Transformation Projects Agriculture.
Food and Optimizing supply chain, improving yield prediction, automating quality control.
Beverage enhancing food safety monitoring Current Perspective on Business Operation 2025, 1.
, 1-20.
Industry Common Benefits of AI-Enabled Organizational Transformation Projects Banking and Financial Fraud detection, risk assessment, customer behavior analysis, personalized financial Services Construction and Real Project planning automation, resource optimization, predictive maintenance of Estate equipment, real-time site monitoring Education Personalized learning paths, intelligent tutoring systems, automated grading, enhanced learning analytics Energy and Utilities Predictive maintenance of infrastructure, demand forecasting, energy efficiency optimization, smart grid management Healthcare Disease diagnosis assistance, medical imaging analysis, patient care improvement, treatment personalization Information Technology Automated network management, cybersecurity enhancement.
AI-driven software and Telecommunications testing, predictive analytics for customer retention Manufacturing Process automation, defect detection, production scheduling optimization, predictive equipment maintenance Marketing.
Advertising Personalized marketing, product recommendation systems, demand forecasting, and Retail customer sentiment analysis Transportation and Route optimization, predictive maintenance of vehicles, real-time tracking, automated Logistics AIAos benefits span the entire organizational value chain, including R&D, maintenance, operations, marketing, sales, production planning, demand forecasting, and service delivery.
Notable outcomes include:
Improved operational efficiency, maintenance, and supply chain management.
Enhanced customer experience and product/service quality through automation and recommendation systems.
Greater adaptability to market changes and support for new business models.
Better forecasting and planning to match supply with demand.
Fraud detection in banking and other sectors.
Automated threat monitoring and information system security.
Advanced diagnostics, treatment, and patient care in healthcare.
From 500 examined cases (Table .
, 40.
6% used integrated AupureAy AI, involving multiple technologies for analytics, task automation, decision support, or digital transformation.
Machine Learning was present in 31.
2% of cases, valued for reducing human intervention and improving system accuracy over time.
Around 18.
6% implemented personal, virtual, or robotic assistants in production processes.
Neural networks, while known for complexity, appeared less frequently, often developed in-house.
Cognitive security AI represented 2% of cases mainly IBM projects, due to its need for extensive complementary resources, diverse AI forms, and vast structured/unstructured data sources.
Current Perspective on Business Operation 2025, 1.
, 1-20.
Table 3.
Industry and Using AI by Type Types of AI .
umber of cases / No.
Industry .
umber of cases / percentag.
IT/telecom/computer services .
/ 14.
80%).
Health .
/ 8.
AI strong or weak .
/ 41.
Bank / financial markets / financial services .
/ 7.
Machine learning .
/ 31.
Business services / professional services .
/ 7.
Deep learning .
/ 2.
Trade / Distribution .
/ 7.
40%).
Automobile .
/ 7.
20%).
Machinery and equipment .
/ 5.
Audiovisual.
Show .
/ 4.
80%).
Electronics .
/ 4.
00%).
Other .
/ 3.
80%).
Metallurgy / Metalworking .
/ 3.
Cognitive .
/ 13%) Cognitive cyber security .
/ 1.
Insurance .
/ 2.
60%).
Research .
/ 2.
20%).
Plastic / Rubber Natural language processing / .
/ 2.
Understanding .
/ 13.
Logistics and transport .
/ 2.
00%).
Education .
/ 2.
Robotic personal assistant .
/ 18.
Pharmaceutical industry .
/ 1.
20%).
Aircraft, rail and shipbuilding .
/ 0.
Petroleum industry .
/ 0.
60%).
E-commerce .
/ 0.
4%).
Aerospace and defense .
/ 0.
Pattern / visual recognition .
/ 2%) Chatbots .
/ 7.
Textile / clothing / headwear .
/ 0.
60%).
Army, security .
/ 60%).
Social .
/ 0.
40%).
Construction / building materials .
/ Neural networks .
/ 0.
Art, design .
/ 0.
40%).
Studies and advice .
/ 0.
20%).
Hotels.
Virtual companion / virtual assistant .
/ restaurants and catering .
/ 0.
Tourism.
Leisure activities .
/ 0.
20%).
Government .
/ 1.
80%).
Agri-food industry .
/ 1.
80%).
Agriculture .
/ 1.
80%).
Energy .
/ 1.
20%).
Communication Ae marketing Ae advertising .
/ Real-time emotion analytics .
/ 1%) Real-time universal translation .
/ 1.
Next GEN cloud robotics .
/ 0.
Autonomous surgical robotics .
/ 0.
Virtual reality .
/ 2.
The case studies also highlight AIAos transformative potential across industries.
In technology sectors (IT, telecom, electronic.
AI strengthens cybersecurity, enhances software development/testing, and supports Big Data-driven business intelligence.
In healthcare and finance.
AI powers computer vision, natural language processing, chatbots, and deep learning to secure sensitive data and improve services.
Manufacturing and construction leverage AI for budgeting, planning, inventory control, resource optimization, and realtime asset tracking.
In transport and logistics.
AI aids predictive maintenance and customer engagement.
Commerce, marketing, and retail benefit from virtual assistants, emotional agents, and chatbots for personalized customer interactions.
Fashion industries apply AI to anticipate trends and tailor shopping experiences, while education uses AI to deliver adaptive, personalized learning paths.
The exploration of the link between IT and organizational performance continues to be relevant, though its focus has evolved over time.
With the rapid advancements brought by artificial intelligence (AI), contemporary research often integrates multiple theories Current Perspective on Business Operation 2025, 1.
, 1-20.
and models to examine how AI influences organizational outcomes and to determine the business value of AI-driven transformation initiatives.
From the literature review, several theoretical underpinnings were identified as suitable for this study: the Paradox of Productivity (Kijek & Kijek, 2019.
Polak, 2017.
Triplett, 1.
, the Process-Oriented Perspective (Mooney et al.
, 1.
, the Resource-Based View (Barney et al.
, 2001.
Grant, 1.
, and the Dynamic Capabilities framework (Kim et al.
, 2.
Insights from these theories were further enriched through a preliminary analysis of the selected case studies to construct a solid conceptual research model.
To ensure the reliability and credibility of the secondary data in the mini-case studies, each was assessed for verifiable details, including: .
the organizationAos contact information, .
the roles, responsibilities, and positions of individuals involved, and .
excerpts from interviews with these actors.
For certain cases, we visited the official websites of the organizations to evaluate the extent of collaboration with their AI solution providers, and in some instances, we contacted organization members directly via the email addresses listed in the reports.
Only case studies with a clearly defined or well-examined issue were included in the final sample.
Each was evaluated based on its context, the challenges addressed, the stakeholders engaged in the transformation process, the results achieved, and its relevance to the research problem.
Guiding questions included: Are the cases comparable? Are the research questions meaningful? Is the data presented in raw form or pre-analyzed? The final sample comprised case studies covering diverse AI applications, such as: AuAIpowered customer query resolution in seconds.
intelligent workflows enhancing insurance customer experiences.
real-time, behavior-responsive smart marketing.
predictive maintenance for machinery and equipment.
image and video recognition in security applications.
personalized financial planning.
fraud detection and antiAemoney laundering systems.
automation of banking and insurance transactions.
customized design and manufacturing.
automated inventory and delivery media archiving and retrieval.
creative content generation in music, marketing, and film.
tailored marketing and advertising.
optimization of supply chains and production systems.
ondemand manufacturing.
self-driving trucks and autonomous delivery.
traffic and congestion management.
improved safety measures.
data-driven diagnostic support.
pandemic detection.
and advanced imaging diagnostics in radiology and pathology.
Ay Table 4.
Hypotesis Summary Hypothesis Results Number of Case Studies AI management capabilities have a significant positive effect on AI capabilities, which are positively associated with the impact of AI at the Supported Supported Supported Supported process level AI personal expertise has a significant positive effect on AI capabilities, which are positively associated with the influence of AI at the process level AI infrastructure flexibility has a significant positive effect on AI capabilities, which are positively associated with the influence of AI at the process level The automational effect has a significant positive impact at the process level, which is positively associated with the influence of AI at the organizational Current Perspective on Business Operation 2025, 1.
, 1-20.
Hypothesis Results Number of Case Studies The informational effect has a significant positive impact at the process level, which is positively associated with the influence of AI at the organizational Supported Supported The transformational effect has a significant positive impact at the process level, which is positively associated with the influence of AI at the organizational level Discussion AI Capabilities Findings from the case study analysis indicate that the effective integration of artificial intelligence (AI) in organizations relies on three core components: .
scalable, high-speed computing infrastructure.
advanced algorithms such as machine learning, deep learning, and neural networks.
extensive, high-quality datasets.
Together, these elements form the foundation of AI capabilities, encompassing AI Management Capabilities (AIMC).
AI Personal Expertise (AIPE), and AI Infrastructure Flexibility (AIIF).
The case evidence shows that organizations typically maintain direct control over AIMC, while AIPE and AIIF are often heavily dependent on external providers.
In certain cases, organizational managers had only a limited understanding of AIAos strategic value, relying on vendors such as IBM or Cloudera to supply both technical expertise and adaptable infrastructure for efficient data processing and value creation.
Notably, many organizations already possessed some form of IT resources, yet these were often underdeveloped, inefficient, or unsuitable for optimal AI deployment.
By combining available resources with automated, intelligent processes supported by robust planning, coordination, and process control.
Organizations were able to enhance decisionmaking quality and streamline workflows.
Regarding AIPE, two distinct scenarios emerged:
On-site collaboration with AI providers Ae Direct vendor involvement facilitated needs assessment, algorithmic process modeling, and system deployment.
This close partnership enabled organizations to reshape strategies, enhance competitiveness, and upgrade products and services.
AI-as-a-Service (AIaaS) or Infrastructure-as-a-Service (AI IaaS) Ae In such arrangements, organizational staff typically did not gain significant exposure to provider Their involvement was limited to defining requirements, specifying data inputs, and using the final solution, with minimal skill transfer.
Overall, these three capabilities can significantly influence organizational processes, driving high performance and maximizing the business value derived from AI transformation Process Level Case studies revealed that AI adoption at the process level is typically aimed at addressing a specific operational challenge, whether through automation, improved information flows, or transformational change.
A deeper examination identified several recurring objectives:
Current Perspective on Business Operation 2025, 1.
, 1-20.
Re-engineering and redesigning organizational structures to enhance customer relationships.
Automating core processes and procedures.
Leveraging diverse internal and external data sources.
Improving the collection, storage, processing, and dissemination of information within and across organizational boundaries.
Restructuring processes to reduce costs, improve integration, and strengthen business intelligence.
Enhancing process efficiency.
Facilitating the acquisition and assimilation of knowledge from both internal and external sources.
Configuring or reconfiguring resources to align with strategic goals.
The informational effects of AI improve administrative performance by enabling faster access to accurate data, enhancing decision quality, and boosting responsiveness in resource management.
The automation effects observed in reduced redundancy, minimized human error, and increased real-time resource visibility, contribute directly to efficiency, reliability, and consistency in operational processes, thereby strengthening both administrative and marketing outcomes.
As a disruptive technology.
AI also generates transformational effects by enabling the creation of innovative products, services, and production processes that address evolving consumer needs.
Examples from the case studies include process re-engineering to enhance customer value, decentralizing operational decisionmaking, raising customer satisfaction levels, and improving product and service quality.
While earlier research often emphasized financial performance as the primary measure of AIAos impact, our findings expand the perspective to include administrative and marketing This broader view underscores AIAos direct and indirect contributions to organizational outcomes and its role in delivering tangible business value through transformation projects.
Conclusions Conclusion The significance of AI has grown increasingly evident as the scope of digital transformation becomes clearer.
Organizations are now recognizing the immense value embedded in the data they possess and the necessity of tools to fully leverage it.
The advancement of AI is driven by two converging forces: the digitization of the economy coupled with the automation of existing processes, and the disruption brought by new service models that capitalize on data utilization.
In this research, which examined the impact of AI on organizational performance, a qualitative method was employed through the analysis of 500 case studies.
This approach was chosen due to its ability to draw meaningful insights from secondary data, enabling a deeper understanding of each case, the extraction of organization-specific details, and the identification of the business value derived from AI transformation initiatives.
It also allowed us to reveal AIAos influence on both organizational and process-level performance.
The findings from this qualitative assessment are summarized in the number of case studies supporting each proposition, as shown in Table 4.
Current Perspective on Business Operation 2025, 1.
, 1-20.
Despite its potential.
AI adoption remains a subject of mixed perspectives.
While many see it as a groundbreaking technological advance with vast potential for organizational growth, its practical applications still face unresolved hurdles.
A key concern is the absence of comprehensive regulations to govern AI operations and ensure ethical compliance (Dignum, 2018.
Hooker & Kim Tae, 2019.
PwC, 2.
Various bodies, including Ie and CERNA, have begun formulating guidelines and issuing recommendations for researchers and developers to create intelligent yet ethical AI systems (Davis, 2005.
Hooker & Kim Tae, 2019.
Schweitzer & Puig-Verges, 2.
The moral implications of AI are also drawing attention, especially in systems relying on deep learning and artificial neural networks, which are often criticized for their Aublack boxAy nature, offering little to no transparency on the reasoning behind algorithmic outputs.
This raises an important trust issue: how can we rely on AI-generated decisions if their underlying logic remains beyond our understanding? Limitation This study has not limitation free.
The information collected is based on secondary rather than primary sources, meaning that data are not captured in real time.
In some cases, datasets may be incomplete or contain gaps.
While the case studies provided substantial insight, there is the potential for bias such as overstated claims or restrictions on publicly shared information.
To enhance accuracy and depth, we recommend using multiple data collection methods, including interviews, surveys, secondary data analysis, and direct observation.
Such triangulation would enable cross-validation and yield richer, more comprehensive findings.
A future replication of this study that incorporates fieldwork, observations, and interviews with relevant organizations could offer additional perspectives and deepen the understanding of AIAos influence on organizational performance.
Furthermore, as this research is cross-sectional, it offers only a snapshot of the relationship between AI and organizational capacity at a single point in time.
It does not account for how these relationships may evolve over extended periods.
A longitudinal approach could track how AI capabilities develop, identify which elements have the greatest impact on long-term performance, and assess productivity trends over One notable limitation of many reviewed studies is the lack of detailed data on the costs and timelines associated with AI-driven transformation projects.
Evaluating these factors could equip organizations with clearer economic justifications for such investments, as well as practical guidance on resource allocation.
Knowing project durations, for instance, would enable the creation of a detailed task schedule that accounts for dependencies and sequencing, thereby improving planning for future AI initiatives.
Implication Every organization requires a clear operational strategy to achieve its goals, which can only be met when leaders mobilize adequate human, technological, and financial resources.
The integration of Artificial Intelligence can address numerous organizational challenges by improving both process-level outcomes .
utomation, information flow, and transformatio.
and overall performance metrics .
inancial, marketing, and administrative result.
This study not only addresses the central question AuCan AI create tangible benefits for my organization, and in what ways can it innovate?Ay but also offers practical guidance for overcoming internal challenges.
To fully leverage AI, organizations should cultivate three critical capabilities:
Current Perspective on Business Operation 2025, 1.
, 1-20.
Mutualization Ae the capacity to adapt existing services for use across multiple A Scalability Ae the ability to handle increasing volumes without compromising system architecture, and A Resilience Ae the capability to maintain operations in the face of disruptions.
Analysis of the case studies suggests that AI adoption is best managed as a project-based initiative, requiring several preparatory and ongoing actions:
A Prepare and train leaders, employees, and stakeholders to understand the specific demands of AI-driven transformation.
A Ensure the quality and sustainability of jobs in humanAemachine collaboration contexts.
A Establish internal and external Aucontrol towersAy to address ethical issues in data and A Recruit and retain AI-specific talent while anticipating shifts in skills, roles, and professional identities.
A Adapt training programs to meet higher volumes and evolving content.
A Adjust governance structures to balance centralized and decentralized decisionmaking.
A Support new AI-enabled workflows with cross-functional collaboration and transparency, while redefining managerial roles.
Unlike many prior studies that focus on a single AI technology in one sector such as healthcare, automotive, or finance.
This research simultaneously examines the use of multiple AI technologies across a wide range of industries.
This holistic perspective reflects the reality that most organizations benefit from a combination of AI solutions rather than a single technology.
For solution providers, the findings underscore the importance of offering integrated AI systems that combine various technologies.
For managers, they highlight the advantages of exploring multi-technology AI strategies to maximize performance gains.
Ultimately, the study emphasizes the multifaceted nature of AI, framing it not as a singular innovation but as an adaptable suite of technological configurations applied across diverse business domains.
This research makes several contributions to the study of AI capabilities (AICAP) and their relationship with organizational performance.
First, it is among the earliest works to assess the direct impact of AICAP on firm outcomes, thereby reinforcing the importance of investing in AI-related expertise and infrastructure flexibility.
Such investments can significantly enhance the business value of AI transformation projects.
Second, it expands the growing body of literature on the organizational benefits of AI, offering a foundation for future research into AI-enabled service innovations and cross-industry applications.
Acknowledgments: The authors would like to express their sincere gratitude to all individuals and institutions that provided administrative and technical support throughout this research.
Special thanks are extended to Universitas Muhammadiyah Purwokerto for their valuable assistance and Conflicts of Interest: The authors declare no conflicts of interest.
The funders had no role in the design of the study.
in the collection, analyses, or interpretation of data.
in the writing of the manuscript.
or in the decision to publish the results.
References