Navigating the AI Frontier 2024

Simple reflex agents operate based on a perception of their environment, without consideration of past experiences.13 Instead, they follow predefined rules to map specific inputs to specific actions. The implementation of condition–action rules allows for rapid responses to environmental stimuli. These early agents are simple rule-based machines or algorithms designed to provide static information and unable to adapt or change course. –Basic spam filters using keyword matching –Simple chatbots with predefined responses –Automated email responders that send prewritten replies following specific triggers Model-based reflex agents are designed to track parts of their environment that are not immediately visible to them.14 They do this by using stored information from previous observations, allowing them to make decisions based on both current inputs and past experiences. By basing their actions on both current perceptions and their internal model, these agents are more adaptable than simple reflex agents even though they are also governed by condition–action rules. Simple reflex agents Model-based reflex agents Goal-based agents Utility-based agents –Smart thermostats that optimize energy usage by adjusting to current and historical temperature data, as well as user preferences –Smart robotic vacuum cleaners that use sensors and maps to navigate efficiently, avoiding obstacles and optimizing cleaning paths –Modern irrigation systems that use sensors to collect real-time data on environmental factors such as soil, moisture, temperature and precipitation, to optimize water dispensation Goal-based agents are able to take future scenarios into account. This type of agent considers the desirability of actions’ outcomes and plans to achieve specific goals.15 The integration of goal-oriented planning algorithms allows the agent to make decisions based on future outcomes, making them suitable for complex decision-making tasks. –Advanced chess AI engines that have the goal of winning the game, planning moves that maximize the probability of success and considering a long-term strategy –Route optimization systems for logistics that set goals for efficient delivery and plan optimal routes by setting clear priorities –Customer service chatbots that set goals to resolve customer issues and plan conversation flows to achieve their goals efficiently Utility-based agents employ search and planning algorithms to tackle intricate tasks that lack a straightforward outcome, thereby going beyond simple goal achievement. They use utility functions to assign a weighted score to each potential state, facilitating optimal decision- making in scenarios with conflicting goals or uncertainty. Rooted in decision theory, this method allows for more advanced decision-making in complex environments. These agents can balance multiple, possibly conflicting objectives according to their relative significance.16 –Autonomous driving systems that optimize safety, efficiency and comfort while evaluating trade-offs such as speed, fuel efficiency and passenger comfort –Portfolio management systems such as robot- advisers that make financial decisions based on utility functions that weigh risk, return and client preferences –Healthcare diagnosis assistants that analyse patient medical records, label patient data (e.g. tumour detection) and optimize treatment strategy recommendations in cooperation with doctorsType Definition Examples Navigating the AI Frontier: A Primer on the Evolution and Impact of AI Agents 11