TABLE OF CONTENTS

AI Is Transforming Tasks, Not Simply Replacing Human Roles

by editor-melon

18 November 2026

TABLE OF CONTENTS

The rapid advancement of Artificial Intelligence (AI) is transforming how businesses operate, manage data, and make decisions. While AI creates opportunities to improve efficiency and productivity, its adoption also introduces new challenges related to information security, data privacy, and accountability.

As AI becomes increasingly integrated into business operations, organizations must ensure that technological innovation is supported by appropriate governance and risk management practices. One approach gaining importance is AI Trust, Risk, and Security Management (AI TRiSM), which focuses on establishing trust, managing risks, and strengthening security throughout AI adoption.

In an episode of Smart ConneXion, Adriansyah Adnan, Chief Technology Officer (CTO) of Anabatic Technologies, discusses how businesses can address emerging AI risks while ensuring that technological innovation remains secure, responsible, and trustworthy.

AI Is Transforming Tasks, Not Simply Replacing Human Roles

One of the most common concerns surrounding AI adoption is its potential to replace human workers. However, technological transformation does not necessarily mean that entire professions will disappear.

According to Adriansyah Adnan, AI primarily replaces or automates specific tasks rather than completely eliminating human roles within an organization.

AI can process large volumes of information and perform operational activities more efficiently. Nevertheless, human involvement remains essential for understanding complex problems, evaluating outcomes, and making decisions that require contextual judgment.

This shift highlights the importance of developing skills beyond routine task execution, including critical thinking, problem framing, and the ability to supervise AI-generated outcomes.

Rather than viewing AI solely as a replacement for human capabilities, businesses should recognize the importance of human-AI collaboration in supporting more effective operations and decision-making.

Why Is AI TRiSM Important for Businesses?

AI technology continues to evolve rapidly, often moving faster than the regulations and governance frameworks designed to oversee its use.

As organizations integrate AI into more business processes, this gap creates challenges in maintaining security, regulatory compliance, and accountability.

AI TRiSM provides an approach to managing trust, risk, and security throughout AI implementation. It emphasizes that successful AI adoption should not be measured solely by technological capabilities but also by the organization's ability to protect sensitive information, control system access, and ensure responsible outcomes.

For businesses, AI governance should be incorporated into the implementation strategy from the beginning rather than introduced after the technology has already been deployed.

ithout appropriate controls, AI adoption may expose organizations to operational vulnerabilities, data privacy risks, and unintended decisions that affect customers and business performance. Establishing clear governance policies is therefore essential to balancing innovation with responsible technology management.

Shadow AI and the Growing Importance of Data Protection

One of the key risks highlighted in the discussion is shadow AI, which refers to the use of AI tools by employees without formal organizational approval, oversight, or compliance with established policies.

Shadow AI may occur when employees use publicly available AI applications to process internal documents, analyze information, or complete work-related tasks without fully understanding how the technology provider handles the submitted data.

This creates potential security and privacy concerns, particularly when sensitive information such as employee records, customer data, or confidential business documents is shared through unauthorized platforms.

In Indonesia, organizations must also consider the requirements of Law No. 27 of 2022 concerning Personal Data Protection (PDP Law), which establishes obligations related to the processing and protection of personal data.

Adriansyah emphasizes the importance of establishing clear corporate AI policies and restricting the use of sensitive information in AI systems.

Techniques such as data masking and anonymization can help reduce exposure by protecting identifying information before data is processed by AI applications.

Beyond technical safeguards, organizations must also strengthen employee awareness and establish approved AI technologies that can be used safely within the corporate environment.

Cybersecurity and the Importance of Security by Design

AI introduces new opportunities and challenges in cybersecurity. While AI can support security operations, it may also be exploited by malicious actors or create additional vulnerabilities within connected business systems.

One emerging concern is prompt injection, where malicious or manipulated instructions influence an AI system to behave in unintended ways.

When AI applications are integrated with internal databases and operational systems, such vulnerabilities may increase the risk of unauthorized access or sensitive information exposure.

To mitigate these risks, Adriansyah highlights the importance of adopting a security by design approach, ensuring that security considerations are embedded from the earliest stages of AI development and implementation.

Organizations should establish clear access restrictions for AI systems, limiting their ability to retrieve information or execute actions beyond their intended functions.

Similar to role-based access controls applied to employees, AI systems should operate with defined identities, permissions, and responsibilities.

This approach allows businesses to maintain greater control over AI integration while reducing potential security vulnerabilities. By incorporating cybersecurity measures from the outset, organizations can create a more secure foundation for enterprise AI adoption.

Human Oversight: Maintaining Control Over AI-Driven Decisions

Despite its advanced processing capabilities, AI is not entirely free from errors or unpredictable outcomes.

For this reason, human oversight remains a critical component of responsible AI implementation. However, Adriansyah explains that simply involving a human in the process does not automatically guarantee effective supervision.

For example, a supervisor who previously reviewed a few dozen documents daily may suddenly be responsible for reviewing hundreds of AI-processed documents. Without adapting the supervision process, increased automation could create additional operational pressure rather than improving efficiency.

To address this challenge, organizations should implement risk-based human oversight, where the level of supervision reflects the potential impact of each AI-driven activity.

Lower-risk tasks, such as summarizing information, may require different controls from activities involving customer interactions or important business decisions.

Organizations can also implement AI guardrails, confidence assessments, and human escalation mechanisms to determine when AI may act independently and when human intervention is required.

In addition, AI systems should support mechanisms for stopping processes, correcting errors, and rolling back actions when necessary.

These safeguards help ensure that human oversight remains practical and effective, even as AI significantly increases processing capacity.

AI Accountability Must Remain with Humans

Another fundamental consideration in AI governance is determining who remains responsible for decisions and actions performed by AI systems.

Although AI can automate tasks and support decision-making, organizations must maintain clear accountability for technology-driven outcomes.

Adriansyah emphasizes that responsibility cannot simply be transferred from humans to AI.

Businesses must establish clear boundaries defining what AI systems are authorized to perform, when human approval is required, and how errors or unexpected situations should be handled.

Individuals or designated teams should retain the authority to supervise AI operations, implement corrective actions, and respond when problems arise. By establishing clear accountability structures, organizations can strengthen governance and ensure that AI-driven activities remain aligned with business objectives, security requirements, and customer expectations.

Building a Secure and Trusted Future for AI Innovation

AI presents significant opportunities for businesses to enhance productivity, streamline operations, and deliver more responsive services. However, realizing these benefits requires more than advanced technology.

Organizations must establish a strong foundation of data protection, access management, cybersecurity, risk-based human oversight, and clear accountability.

Through an AI TRiSM approach, businesses can pursue technological innovation while maintaining the controls necessary to manage emerging risks.

Ultimately, the future of AI is not defined solely by technological advancement, but also by how effectively organizations ensure that AI remains trustworthy, secure, accountable, and beneficial to their business.

Watch the full discussion in Smart ConneXion featuring Adriansyah Adnan, Chief Technology Officer (CTO) of Anabatic Technologies, to explore how organizations can address AI risks and build trusted innovation.

Discover more about KPSG's technology and business transformation solutions at www.kpsg.com or contact us at https://kpsg.com/en/contact-us/

Other insights

banner-headset-center
The Next Chapter of Indonesia’s Tech Industry: AI, Human Roles & Trusted Innovation
gen AI