2025 Agentic AI: When AI can think and make decisions on its own.
In 2024, the rise of Gen AI significantly impacted the workplace. Users could create new things from existing data, such as writing text, images, audio, video, or even writing software programs, simply by writing prompts that provided the desired results. This made work easier, generated revenue, and reduced costs.
This year, the concept of Agentic AI is gaining popularity. This concept proposes AI that can work independently, make decisions, execute actions, and even learn on its own. Agentic AI is being closely watched for its potential development and widespread implementation by the end of this year or early 2026.
What is Agentic AI?
Agentic AI is an AI that can use received data to make decisions, take action, and complete assigned tasks independently. Agentic AI can also analyze data and work autonomously, learning and adapting its work methods to complex environments.
The difference between Gen AI and Agentic AI.
Generative AI will follow requests or commands it receives, such as asking and answering questions, writing articles, creating images, or developing code, but it cannot do anything beyond those commands. For example, if asked to find stores selling IP phones priced between 2,000 and 3,000 baht in the Chatuchak area, the Gen AI will quickly find the answer.
Agentic AI is capable of handling more complex requests, such as making decisions or performing actions to achieve desired results. For example, once a store in Chatuchak selling IP phones priced between 2,000 and 3,000 baht is found, the Agentic AI can immediately arrange for the purchase and delivery of the goods to the company.
Why can agentic AI perform complex tasks?
Agentic AI operates using four key structural components:
- Perception – Gathering information from multiple sources, such as using sensors to read and input data from documents or databases; processing data from various sources accurately and quickly to understand the surrounding situation .
- Reasoning ( Analysis and Decision Making) – Uses advanced algorithms to analyze and evaluate problems and decide on the most appropriate approach in a step-by-step manner. Can handle complex problems.
- Action – Working towards clear goals and executing planned actions without waiting for instructions from humans.
- Learning – The ability to analyze oneself, identify mistakes, improve and adapt, and learn from feedback and experiences.
The application of Artificial Intelligence (AI) in various businesses.
Contact center
- Intelligent AI Voice Bots and Chatbots – analyze customer emotions and behavior to respond appropriately without human intervention.
- In-depth analytics – Use AI to evaluate conversational data to detect trends in business problems or opportunities.
- Automated customer tracking system – helps notify customers about special offers, service renewals, or after-sales service without requiring human labor.
finance
- Analyze investment risks and make investment decisions in the stock market automatically.
Robots and automation
- Intelligent robots that can adapt to situations, such as self-driving cars that can decide to change routes based on traffic conditions.
insurance:
- They can assess the damage and decide whether to approve compensation.
telecommunications:
- It can immediately monitor and resolve telecommunications network problems without waiting for human intervention.
public health:
- It can help in automatically analyzing a patient’s symptoms and recommending appropriate treatment.
Benefits of using Agentic AI
- Improve work efficiency – speed up processes, enable repetitive tasks to be performed continuously, and reduce human workload.
- Learning and adapting – the ability to continuously improve one’s performance and adjust to changes or new user needs.
- Reduce errors – Analyze large amounts of data quickly, leading to more accurate decision-making and reducing human error.
Things to consider when implementing Agentic AI.
⚠️ Data Privacy – Agentic AI needs access to sensitive user data to provide effective personalized recommendations. Therefore, the storage, processing, and compliance with data privacy laws such as GDPR and PDPA to protect data securely are essential.
⚠️ Ethics – When agentic AI can make its own decisions, ethical questions may arise, such as transparency in the work process, the integrity of the algorithms used for decision-making, and the impact on humans from those decisions.
⚠️ Complexity in development – More advanced technology is required for development to ensure that Agentic AI can operate independently, efficiently, and reliably.
The future of AI
It is certain that AI will play a significant role in many industries, especially in the development of AI agents or digital workers to become important assistants for humans. In the near future, we will see AI agents that help manage request forms in after-sales service, AI agents that help search for airline tickets and hotels and manage bookings with the best deals, or AI agents that assist in negotiating disputes and insurance claims.
In the future, human work will inevitably involve collaboration with AI agents. Therefore, humans must adapt, transform themselves, and develop their skills and abilities to maximize their potential and work effectively and smoothly with AI agents.
