Unlocking transformative innovations for future industries with Agentic AI

This article is the second part of a four pieces series on the rise of AI agents. You find find part one here.

6. Current AI agents breakthroughs

The proliferation of sophisticated AI agents 

Building on the fundamental knowledge of AI agents, this section emphasises their growing complexity and practical use in a variety of industries, supported by current data and tangible examples. The increased capacity of AI agents to comprehend complicated contexts, solve difficult issues, and carry out activities with more autonomy is a defining characteristic of their evolution.

Healthcare

AI agents are having a big impact on the healthcare sector by increasing patient care, accuracy, and efficiency.

• Diagnostic support and medical imaging: With accuracy on par with or better than human radiologists, AI agents can evaluate medical pictures such as X-rays, MRIs, and CT scans to find abnormalities. AI systems are being utilised, for example, to detect Alzheimer's, cardiovascular disease, and cancer early. By learning from large collections of medical imaging data, these agents are able to spot small patterns that may indicate a disease. They frequently flag photos for expert assessment, increasing the accuracy of diagnosis and possibly enabling early treatment

• Personalised treatment and predictive analytics: AI agents can customise therapy routes by combining multiple data sources, including a patient's genetics, lifestyle, and medical history. They can forecast how patients will react to various medicines as well as possible side effects, which helps doctors better customise care. AI helps determine the best chemotherapy regimens in cancer, for instance, by balancing effectiveness with the least amount of adverse effects

• Virtual health assistants and patient monitoring: Through wearable technology, AI-powered virtual assistants provide ongoing patient support by responding to questions, reminding patients to take their medications, and keeping an eye on vital signs. By identifying early warning indicators of serious illnesses like sepsis or heart failure, these agents can reduce emergency hospital visits and notify medical professionals for prompt intervention. AI agents, for example, are being utilised to improve independent living and provide carers peace of mind by providing elderly patients with companionship and health check-ins

• Administrative automation: Electronic health record (EHR) updating, claims processing, and appointment scheduling are among the administrative duties that AI agents are expediting. Healthcare workers may concentrate on providing direct patient care by using AI to automate these repetitive processes, which also lowers administrative workloads and boosts overall operational effectiveness. AI-powered automation can drastically cut down on the amount of time spent on tasks like processing insurance claims and entering patient data, according to studies

Manufacturing

AI agents are revolutionising manufacturing processes, leading to increased efficiency, improved quality control, and reduced downtime.

• Predictive maintenance: Machine sensor data is analysed by AI agents to anticipate possible equipment breakdowns before they happen. These agents enable preventative maintenance by spotting minute patterns in temperature, vibration, and other variables, minimising unplanned downtime and lowering repair expenses. Predictive maintenance powered by AI, for instance, has been demonstrated to cut downtime in some manufacturing environments by as much as 40%

• Quality control: On production lines, AI-powered visual inspection agents are used to find flaws more quickly and precisely than human inspectors. These agents, who have been trained on certain product standards and defect typologies, examine data and photographs in real time to guarantee consistent product quality and lower the quantity of faulty products that reach customers. Following the implementation of AI-powered visual inspection, one automaker reported a 90% decrease in problematic components

• Production planning and scheduling: To optimise production schedules, AI agents examine inventory levels, demand projections, and production capacities. These agents assist producers in cutting excess inventory, shortening lead times, and increasing overall efficiency by dynamically modifying production schedules based on real-time data

• Supply chain optimisation: Large volumes of data, including logistical data, market trends, and supplier performance, are tracked and analysed by AI agents across the supply chain. This helps producers lower costs, minimise risks, and guarantee on-time delivery of goods by facilitating more intelligent purchasing and logistical decisions

Customer service

AI agents are transforming customer interactions by providing 24/7 support, personalised experiences, and efficient issue resolution.

• AI-powered chatbots and virtual assistants: Advanced AI agents are able to retrieve data from various internal systems, comprehend natural language, and carry out operations like appointment scheduling, record updating, and refund processing. Wait times can be decreased and customer satisfaction raised by these agents' ability to manage a high volume of consumer enquiries concurrently. As an illustration, businesses that have implemented chatbots driven by AI have claimed notable decreases in handling times and improvements in customer engagement

• Personalised recommendations and support: In order to offer individualised product suggestions and support, AI agents examine previous consumer interactions and preferences. These agents can provide customised solutions and improve the entire customer experience by knowing the demands of each unique consumer

• Sentiment analysis: Businesses can gain real-time insight into customer sentiment by using AI agents to evaluate client interactions and detect emotional tones. This makes it possible for them to give priority to pressing problems, respond sympathetically, and raise client satisfaction levels all around

• Automated ticketing and routing: By automatically classifying and forwarding consumer enquiries to the relevant human agents, AI agents can expedite the settlement of complicated problems and boost the effectiveness of the customer support staff

These instances highlight the significant effects of increasingly complex AI agents across a range of industries, and they are backed by new data and continuous developments. Significant gains in productivity, efficiency, and overall results are being driven by their capacity to learn, adapt, and act independently, opening the door for even more revolutionary uses down the road.

Table 1: Summary of breakthrough areas

7. The transformative potential of agentic AI for ASEAN and Asia

The ascent of agentic AI

This section concentrates on the new paradigm of agentic AI, whereas the previous section demonstrated the broad application of advanced AI agents for certain tasks. This is a big step forward, as AI agents develop from task-oriented instruments to increasingly independent beings who can solve complicated problems, learn from mistakes, and actively pursue more general objectives without continual human assistance.

Key characteristics of agentic AI

• Autonomy and proactivity: Agentic AI has the ability to autonomously determine objectives, devise plans of action, and start actions to accomplish them, in contrast to standard AI agents that mostly respond to certain inputs.

• Reasoning and planning: These sophisticated agents can evaluate complex situations, create multi-step plans, and modify their tactics in response to feedback and evolving conditions because they have improved reasoning abilities.

• Learning and adaptation: Without intentional reprogramming, agentic AI gradually improves its performance and expands its knowledge through interactions and experiences.

• Memory and contextual awareness: They exhibit a greater level of contextual comprehension by being able to remember details from previous encounters and apply them to guide judgements in the future.

Software development and it operations

Agentic AI is beginning to automate significant portions of the software development lifecycle and IT management.

• Autonomous code generation and refinement: With the help of high-level specifications, AI agents can now produce large quantities of code and even optimise and modify existing codebases on their own. For example, GitHub Co-pilot shows how AI can greatly speed up coding chores, even though it still needs human assistance. It is anticipated that subsequent iterations will demonstrate increased independence in comprehending project objectives and producing more comprehensive solutions

• Automated incident response and remediation: Agentic AI is being developed for use in IT operations to track system performance, identify irregularities, identify underlying causes, and automatically apply fixes to restore services without the need for human participation. Businesses are looking into AI-powered systems that can learn from previous problems and proactively avoid new ones by smartly allocating resources and modifying system configurations

Scientific research and discovery

By independently planning and carrying out experiments, assessing data, and developing new theories, agentic AI is quickening the rate of scientific discoveries.

• Autonomous laboratories and experimentation: Robotic labs with AI capabilities that can independently carry out intricate experimental procedures, gather and analyse data, and refine experimental designs are early examples. Continuous operation of these systems can speed up the scientific procedure and possibly result in quicker advances in materials research and drug development

• Hypothesis generation and knowledge synthesis: Large volumes of scientific literature can be processed by agentic AI, which can then find connections and patterns that human researchers might overlook and produce original theories for more study. This capacity could speed up the comprehension of intricate scientific phenomena and open up new study directions

Financial services and trading

Agentic AI is being deployed for sophisticated financial analysis, algorithmic trading, and risk management.

• Autonomous trading agents: With less human intervention, sophisticated algorithmic trading systems are becoming more autonomous and able to make intricate trading decisions based on risk assessment, real-time market information, and predetermined investment plans. Compared to traditional systems, these agents are able to recognise lucrative possibilities and adjust to shifting market conditions more quickly

• Intelligent risk assessment and compliance: Large datasets can be analysed by agentic AI to find possible threats, spot fraud, and guarantee adherence to intricate legal requirements. These systems can improve the security and stability of financial institutions by learning from previous fraud cases and proactively modifying their detection techniques

Logistics and supply chain management

Agentic AI is optimising complex logistical operations and supply chains with minimal human intervention.

• Autonomous fleet management and route optimisation: AI agents are being created to autonomously manage car fleets, maximising routes in real time according to delivery schedules, traffic conditions, and other dynamic variables. This results in lower transportation expenses, quicker delivery, and increased effectiveness

• Intelligent warehouse management: With little human guidance, agentic AI can manage and optimise warehouse operations, such as order fulfilment, inventory control, and robotic systems, maximising throughput and responding to shifting demands

The trajectory towards artificial general intelligence (AGI)

Improvements in autonomy, reasoning, and learning capabilities are seen as essential steps towards the long-term objective of Artificial General Intelligence (AGI) – AI with human-level cognitive abilities across a wide range of tasks – even though current agentic AI systems are still domain-specific. The advancement of increasingly complex agentic AI offers important new information and technical foundations that may someday help bring AGI to fruition.

Agentic AI is no longer a far-off theoretical idea; rather, it is a quickly developing reality that has the potential to completely transform a wide range of industries, as evidenced by the instances and trends discussed in this section. Their influence on the world economy and society will only increase as these agents develop further and gain the ability to act and learn on their own, especially in dynamic regions like Asia and ASEAN.

Table 2: Potential applications of agentic AI in ASEAN and Asia

After shedding light on the most recent developments in AI agents and the exciting potential for transformation in a variety of ASEAN and Asian industries, the following section of this commentary (Part 3) will turn our attention to the fundamental tactics needed to responsibly utilise this new power. As crucial pillars for the region's successful navigation of the agentic AI revolution, we will examine the imperatives of funding research and development, creating strong ethical and regulatory frameworks, and cultivating productive public-private collaborations.

illuminem Voices is a democratic space presenting the thoughts and opinions of leading Sustainability & Energy writers, their opinions do not necessarily represent those of illuminem.