By: Tayyab Kaleem Khan
The wave of advancements in artificial intelligence has taken the industry by storm. Some pose it as the fifth industrial revolution, marked by the most transformative technological shift of our time. From AI agents managing patient appointments to autonomous defense systems and predictive policing, AI is becoming an integral part of our daily lives. Its diffusion is bidirectional, both horizontal, with an overarching impact across sectors and vertical, specific to industries. However, with the AI proliferation, there comes the challenge of ensuring end user safety, ethics and accountability.
AI adoption has transcended boundaries and is no longer limited to the North American continent. With Asian companies like DeepSeek and Manus leaping forward, the technology is being adopted rapidly, and its impact is being felt. Startups and small enterprises have started deploying their own agents due to the availability of open-source models (e.g., DeepSeek R1), accelerating democratization. However, with the advancement comes the perils of its misuse. It has been reported that cyber-attacks and deep-fake agentic AI voices and images have increased financial frauds as well. With this rapid adoption, the pace to govern it is falling behind and the need for implementation of guardrails to protect the society is ever increasing because of the dual use nature of the technology.
The technological frontier is promising, as a report by PWC predicts that AI will contribute US$ 15.7 trillion to the global economy by 2030, with $6.6 trillion coming from enhancements in productivity and US$9.1 trillion originating from the demand side. At this critical juncture, Asia stands to gain the biggest portion of the pie, China alone has an expected market of US$ 7 trillion and North America will have US$ 3.7 trillion. Mckinsey Global Institute has estimated that AI can contribute to 1.2% growth of GDP per year for the next decade.
Despite the positives, the AI revolution comes with social and economic costs. There will be immense labor force pressures as the AI agents proliferate all levels of skill-based jobs and replace them. World Economic Forum has estimates that over 85 million will lose jobs within 2025 whereas additional 97 million jobs could emerge if its adapted well for supplementing human activity.
In Asia, the highest impact is projected to be in the manufacturing and retail sector which needs to be prepared for this transformation. Whereas the areas where there can be potential growth are AI ethics, data science and digital health. However, this transition is unlikely to be smooth unless supported by proactive investments and policy making.
As evident from data, the potential risks are real. The large corporations are better poised for adoption due to their ability to deploy capital to embrace AI technology, whereas the smaller players risk extinction such as call-centers or IT services industry which impacts a big chunk of businesses in countries such as Philippines, Pakistan, and India. However, this diffusion must be coped up by the regulators to deliberate that the adoption is met with governance mechanisms to ensure the adopted AI models have oversight. In countries with low regulatory infrastructure, players will move towards deployment of unchecked AI systems with minimal scrutiny which can put the society at risk of bias, misinformation, and lead to general erosion of public trust.
To address these challenges, there should be a strategy to ensure responsible use of artificial intelligence, The AI adoption must be under the premise of guardrails that evolve alongside its diffusion. Meaning, these guardrails come in several forms: regulatory, technical and institutional.
First, Regulatory frameworks are essential, but they must be adaptable. The European Union’s AI Act is one of the most comprehensive efforts to classify AI systems by risk and assign regulatory obligations accordingly. Yet it also reveals the difficulty of crafting one-size-fits-all policies in a rapidly changing landscape. In regions like Asia, fragmented approaches often lead to regulatory arbitrage or inconsistent protection for users.
Secondly, the technical guardrails are standardization, specific practices and frameworks guiding the design of artificial intelligence systems to ensure they operate safely, ethically, and effectively. These guardrails encompass various aspects of AI development and deployment and play a pivotal role. This includes principles of explain-ability, robustness, and fairness which must become essential criteria for practical implementation rather than mere theoretical aspirations.
As spread of AI technologies become increasingly pervasive, it is important to develop systems that are transparent in their decision-making processes and subject to thorough audits after deployment. Approaches like differential privacy, federated learning, and adversarial testing must be integrated into standard practices to enhance the resilience of AI models against attacks and to maintain data integrity.
Lastly, the institutional guardrails refer to the governance structures within organizations, and independent watchdogs to act as a mechanism of checks and balance.
In organizations internally, the leadership and management boards, especially in large companies deploying AI at scale, need to fully grasp the associated risks and ensure oversight that extends beyond engineering teams.
Similarly, forming independent AI ethics councils, interdisciplinary advisory groups, and effective whistle-blower protections can establish necessary internal checks and balances. By ensuring that AI development aligns with long-term societal values, these governance frameworks can help prevent short-term pressures from overshadowing ethical considerations.
The proposed approach and call to action for guardrails may be perceived as red-tapism but a well-defined standards and transparent governance can enhance trust, reduce resistance, and create more stable markets for AI solutions. For instance, in the financial sector, organizations that adhere to principles of explain-ability and fairness are more likely to gain institutional acceptance. Likewise, in healthcare, AI tools that are rigorously validated and audited are more likely to find their way into national health systems.
To sum up, guardrails provide the reliability and trust needed for enduring innovation. It is important to understand that guardrails should not be viewed as impediments to innovation. The spread of AI is inevitable, the direction it takes is not fixed. We are at crossroads where the choices made by developers, regulators, and society will shape whether AI acts as a force for inclusion and empowerment or as a catalyst for division and erosion of freedoms. Therefore, building strong and adaptable guardrails alongside the diffusion of AI is not simply a regulatory task; it is a societal necessity.
About the author: Tayyab Kaleem Khan is a Management Consultant in Public Sector.
