As we look forward to the 2024-2025 school year, it's clear that AI will play an increasingly critical role in education. Having spent the past 18 months studying AI's impact on teaching and learning, here are my top five predictions for how AI will transform schools in the upcoming school year. 

1. Advanced AI Agents

2. Enhanced Voice Assistants

3. Multimodal AI Models

4. Improved AI Personalization

5.  AI as Customer Service Agent

#1 Advanced AI Agents

What Are AI Agents?

AI agents are advanced artificial intelligence systems designed to perform specific tasks autonomously without continuous human intervention. Unlike traditional AI tools that require user input for each task, AI agents can operate independently, make decisions, and complete tasks based on their programming and learned experiences.

How Do AI Agents Differ from Current AI Tools?

• Autonomy. Most AI tools today (e.g., EasyGrader and Gradescope) require frequent human input and oversight. For example, a teacher might use a grading tool that automates parts of the grading process but still needs to manually input data or review results. AI agents will be able to manage the entire grading process, from collecting assignments to providing feedback and recording grades.

• Complexity of Tasks. Current AI tools often handle specific, straightforward tasks, such as voice-to-text transcription or basic data analysis. AI Agents will be capable of performing more complex and multi-step tasks. An AI agent in education might not only grade assignments but also analyze student performance trends over time, identify areas where students are struggling, and suggest personalized resources to help them improve.

• Learning and Adaptation. Many existing AI tools have limited ability to learn from past interactions. They perform well within their defined scope but don’t adapt significantly over time. AI agents can learn from their interactions and improve their performance based on feedback and new data. In a school setting, an AI agent might learn a teacher’s preferences for grading or adjust its support based on the unique needs of different classes.

• Integration and Coordination: AI tools today typically function in isolation and are used to complement human tasks. AI agents in the near future will integrate with various systems and coordinate multiple tasks simultaneously. For example, an AI agent might help take attendance and communicate the absences to a student information system (SIS).

In the end, AI agents represent a more advanced and autonomous form of AI technology that can perform complex tasks independently. For educators, this means less time spent on administrative duties and more time available for instructional activities and student interaction. Of course there are challenges that will need to be addressed, but the technology will soon be available for those who want it. 

#2 Enhanced Voice Assistants 

What Are Enhanced Voice Assistants?

Enhanced voice assistants are advanced versions of the current voice-activated AI tools we have today. These assistants, such as those developed by Google and Amazon, are designed to understand and respond to voice commands more accurately and intuitively. They leverage improvements in natural language processing (NLP) and machine learning to offer more personalized and context-aware interactions.

How Do Enhanced Voice Assistants Differ from Current Technologies?

• Improved Understanding and Responsiveness. Existing voice assistants can perform basic tasks like setting reminders, playing music, or answering simple questions. However, they often struggle with complex queries or commands, especially those that require context or nuanced understanding. The next generation of voice assistants will be much better at understanding context and providing accurate, relevant responses. They will be able to handle multi-step requests and understand natural conversation which will make interactions smoother and more effective.

• Context-Awareness. Today's voice assistants have limited context-awareness. They can remember certain details within a single session but often lose context when the conversation gets more complex or spans multiple interactions. Enhanced Voice Assistants will be able to retain context over longer interactions and across different sessions. They will remember user preferences, past interactions, and even ongoing projects which will provide for more personalized and relevant assistance.

• Personalization. While current tools offer some level of personalization, it's often basic and limited to pre-set preferences or simple user data. Future voice assistants will use advanced machine learning algorithms to tailor their responses and actions to each user's unique needs and preferences. This means they can adapt to different learning preferences, assist with personalized study plans, and offer support based on individual student progress.

• Integration with Educational Tools:  Integrations for current tools are minimal and often requires additional setup, and many have compatibility issues. Enhanced voice assistants will seamlessly integrate with a wide range of educational technologies, from classroom management software to learning management systems (LMS). 

Enhanced voice assistants will significantly elevate the capabilities of current AI voice tools and provide more intuitive, responsive, and personalized interactions. For teachers, this can mean better classroom management, more tailored support for students, and innovative ways to assist students with disabilities. It’s my hope that by integrating these advanced voice assistants into their daily routines, teachers can create a more efficient, inclusive, and engaging learning environment for their students.

#3 Multimodal AI Models 

What Are Multimodal AI Models?

Multimodal AI models are advanced artificial intelligence systems capable of processing and generating content across multiple types of data— text, images, audio, and even video. These models will be able to understand and synthesize information from various modalities to create richer and more interactive outputs.

How Do Multimodal AI Models Differ from Current AI Tools?

• Integration of Multiple Data Types. Most existing AI tools specialize in a single type of data. For example, language models like GPT-4 process text, while image recognition models handle visual data. Multimodal AI models will be able to process and generate content that spans text, images, audio, and video, and allow for a more comprehensive and integrated approach to problem-solving and content creation.

• Enhanced Learning Experiences. Learning tools today are often limited to one form of media, such as text-based learning platforms or audio-based language apps. Multimodal AI models will be able to combine text, images, and audio to create more immersive and interactive learning experiences. For instance, a lesson on the solar system could include text descriptions, 3D visualizations of planets, and audio narrations.

• Interactive Simulations. Existing tools might offer basic simulations or visualizations, but they’re often limited in scope and interactivity. Future models will enable highly interactive simulations that can be manipulated in real-time, and provide students with hands-on learning experiences in any virtual environment.

Multimodal AI models will revolutionize the way we approach teaching and learning by integrating visual, auditory, and textual information into cohesive and interactive lessons and activities. For teachers, this means the ability to create more engaging and effective lessons that cater to diverse learners and provide deeper understanding through immersive simulations and visualizations. Embracing these advancements will not only enhance subject comprehension but also create a more dynamic and interactive classroom environment for students at every grade level. 

#4  Improved AI Personalization

What Is AI Personalization?

AI personalization refers to the ability of artificial intelligence systems to adapt and respond to individual users based on their unique preferences, behaviors, and contexts. These systems can learn from past interactions, remember user-specific details, and provide customized experiences that cater to individual needs. We’re not too far from this prediction today.

How Does Improved AI Personalization Differ from Current Technologies?

• Enhanced Memory and Context Awareness. Existing AI systems have limited memory and context-awareness. They often forget user preferences between sessions or fail to fully adapt to individual needs. Advanced AI systems will retain user preferences and contextual information over time, and offer more consistent and personalized interactions. 

• Customized Learning Experiences. Many educational tools offer a one-size-fits-all approach, with limited customization based on individual student performance and preferences. Future AI systems will tailor learning experiences to individual student needs, identify strengths and weaknesses, and provide targeted support. This could involve personalized lesson plans, adaptive quizzes, and customized feedback to help each student achieve their best.

• Real-Time Adaptation. Adaptation in current tools is often slow and requires significant data input before changes are made. Advanced AI systems will adapt in real-time to user inputs and behaviors. For instance, if a student struggles with a particular concept, the AI will be able to offer additional resources or alter the teaching approach to better suit the student’s needs.

• Holistic Student Profiles. Profiles and data in current tools are often fragmented across different systems. This makes it challenging to get a comprehensive view of a student’s learning profile. Enhanced AI systems will integrate data from multiple sources to create holistic student profiles. These profiles will be able to track progress, preferences, and areas for improvement across different subjects and over time, provide a more complete picture of each student’s learning progress. 

Imagine a classroom where each student receives a tailored educational experience. With improved AI personalization, this is not out of the question. For teachers, this means being able to address the diverse needs of their students more effectively, and being able to ensure that every student receives the support and challenges they need to be successful. 

#5 AI as Customer Service 

What Are AI-Powered Customer Service Agents?

AI-powered customer service agents are sophisticated artificial intelligence systems designed to handle customer interactions autonomously. These agents will manage inquiries, process transactions, and provide support across various communication channels, such as email, chat, and phone.

How Do AI-Powered Customer Service Agents Differ from Current Technologies?

• Efficiency in Handling Inquiries. Traditional customer service often requires human intervention for managing inquiries and providing support. AI-powered agents will be able to respond to inquiries instantly, and provide information and resolve issues without the need for human involvement. This will ensure parents, students, and teacher inquiries will be handled faster, more efficiently, and more accurately. 

• Consistency and Availability. Human-operated customer service (e.g., front desk reception and technology support) is limited by office hours and can vary in response time. AI agents will be available 24/7 and be able to provide consistent and reliable service. They’ll also be able to handle a high volume of interactions simultaneously and will be able to do it without compromising quality in any way. 

• Advanced Interaction Capabilities. Basic automated systems can handle simple tasks but struggle with complex queries or multi-step processes. Advanced AI agents will manage complex interactions, understand context, and provide detailed support. They’ll be able to handle tasks like processing applications, answering detailed policy questions, and managing technology-related questions.

• Data-Driven Insights. Current AI tools often rely on limited custom data and may lack deeper insights into common issues or specific user needs. AI systems in the coming months will be able to analyze large volumes of custom data and identify trends and common issues.

AI-powered customer service agents will significantly enhance the efficiency and effectiveness of school communications. By managing parent inquiries, human resource services, and administrative support autonomously, these agents will ensure timely and consistent support. For administrative teams, this means less time spent on repetitive communications and more time available for direct engagement with students, teachers, and parents. 

As we prepare for the 2024-2025 school year, the potential of AI in education is undeniable. From autonomous AI agents to personalized learning experiences, these advancements promise to reshape the way we teach, learn, and interact within the educational ecosystem. By embracing these innovations, we can streamline administrative tasks, enhance classroom engagement, and provide tailored support to meet the diverse needs of every student. As we move forward, it will be essential to stay informed, adapt to unpredictable changes, and be open to AI's capabilities as we all work to create a more efficient, inclusive, and impactful educational experience for our students.