For decades, people have debated how to use technology to revolutionize education, whether by “gamifying” instructional materials or expanding access to knowledge through massive open online courses.
Schools spent nearly $160 billion on education technology, or ed-tech, in 2016, according to EdTechXGlobal and Ibis Capital, and spending is expected to grow at a 17 percent annual rate through 2020. From 2011 to 2015, private investment in educational technology increased by 32% annually, reaching $4.5 billion globally.
The contribution of AI to these flows has not been calculated. Nonetheless, it is likely to rise as artificial intelligence technologies are well suited to achieving important educational goals such as improving teaching efficiency and effectiveness, providing education to all, and developing 21st-century skills.
So, in terms of artificial intelligence, where will education be in 2030? It will almost certainly play a significant role. On the other hand, success depends on technical and ethical issues, beginning with who owns student data, who can see it, who can use it, and for what purposes.
Bridging the skills gap
Many countries suffer from significant skill mismatches, which are caused by the education system’s inability to accurately reflect employer demands and labor market frictions that prevent individuals from being properly matched to jobs. Only half of the students in a survey of ten developed and developing countries believed their post-secondary education improved their employability. More than a third of employers cited skills shortages as a major reason for entry-level job openings. The resulting skills gap not only causes economic underperformance, but it also prevents many people from reaching their full potential.
Artificial intelligence will also play a key role in improving the connection between education and labor markets. By connecting talent with job opportunities, digital technologies are already making a difference. According to a recent MGI study, by 2025, online talent platforms could help up to 60 million people find work that better matches their skills or preferences while also lowering the cost of human resources management, including recruitment, by up to 7%. Artificial intelligence’s opportunities in employment-to-education settings have already begun to attract new players, thanks to a growing emphasis on lifelong learning.
Improved pattern recognition enabled by machine learning and detailed data on potential employees may help to improve recruitment results in the future. It can help hiring companies pinpoint the exact skill sets and personality traits that will enable someone to succeed in a job and uncover previously untapped insights in talent management. Artificial intelligence could also help recruiters avoid using school reputation as a proxy for evaluating candidates’ potential by detecting promising candidates with less traditional credentials. Fundamentally, artificial intelligence will improve education systems’ ability to meet the needs of future employers.
Attracting students and keeping them
Educators will be able to use personal, academic, and professional data and government data to ensure that students benefit from the courses they choose. The value is derived from students’ ability to excel academically and the institutions’ ability to assist them in finding meaningful employment. People who appear unsuitable based on traditional measures of academic success but have high potential based on other abilities and traits could be identified using machine learning. Students will benefit from better targeting because it will allow institutions to attract the right mix of people, improve learning outcomes, and help schools and universities improve their offerings over time.
Universities are already looking into AI applications to help students stay in school longer. Some colleges and universities are experimenting with advanced analytics and machine learning to identify students with difficulties and assist them before they drop out. Civitas Learning and Salesforce have teamed up to create a service for universities that identifies and engages students who are on the verge of dropping out. Machine learning is used by Salesforce tools to recommend engagement strategies that improve retention and graduation rates.
By monitoring students as they work, tracking their eye movements, and observing their expressions to see if they are engaged, confused, or bored, computer vision could detect signs of disengagement in the future. Some institutions in the United Kingdom are experimenting with computer vision, natural language processing, and deep learning algorithms to better understand students’ learning difficulties and preferences, incorporating novel data types such as students’ social media activities.
Unleashing personalized learning
Attracting and retaining students is critical, but the real educational breakthrough will most likely come from a fundamentally different approach to learning, whether in or out of the classroom. In recent decades, many efforts have been made to tailor learning to each student and move away from a standardized approach. Adaptive learning solutions seek to overcome the limitations of traditional classroom instruction by tailoring lesson plans to a student’s prior knowledge, learning preferences, and progress. Adaptive learning claims to deliver the right content, at the right time, in the best way to each student, rather than delivering a single lesson to the entire class, which can leave struggling students behind or disengage fast learners.
Artificial intelligence could improve adaptive learning and personalized teaching by identifying factors or indicators of successful learning for each student that was previously impossible to capture. In addition to tracking variables like the number of times a student pauses during a lesson, the amount of time it takes to answer a question, and the number of times a question is attempted before getting it right, computer vision and deep learning could pull in additional data like mouse movements, eye tracking, and sentiment analysis, providing deeper insights into a student’s performance, confidence, mindset, and cognitive ability.
AI-enabled adaptive learning could restructure education if implemented at scale. It could do away with traditional testing systems in favor of a more nuanced assessment of academic abilities and achievement. Teachers would focus less on lecturing and more on coaching, aided by prescriptive analytics to choose the most effective methods. Class formats would give students more room to learn according to their preferences, with teachers focusing less on lecturing and more on coaching.
Releasing teachers’ true value add
Teachers’ jobs may be stripped of time-consuming administrative tasks in the future, such as supervising and answering routine questions. Teachers would have more time to mentor and coach students, which are valuable tasks uniquely suited to humans.
Natural language, computer vision, and deep learning could help students with routine questions or as tutorial supervisors, allowing teachers to focus on other tasks. A virtual supervisor could use AI to track students’ work and behavior and provide teachers with statistically-based insights and constructive feedback on their progress. In the future, AI solutions may be able to monitor an entire classroom and call out students individually using voice and facial recognition.
Finally, by applying machine learning algorithms to data from students’ education profiles, social media, and surveys, AI in education could assist teachers in forming the most effective groups or classes. Companies like Collaboration.ai use artificial intelligence to process data on each student’s experience, knowledge, and capabilities, create instantaneous maps of connections and networks, highlight each student’s unique potential, break down preferences and bias, and recommend best-suited group formations for the learning objective. Complementary skills that maximize critical thinking and test students’ ability to adapt and collaborate can be identified using machine learning.
Toward virtual teachers
According to UNESCO, the world will need to recruit and train 24.4 million primary school teachers by 2030 to achieve universal primary education and 44.4 million secondary school teachers to fill openings by 2030. Many of these new hires—more than 85% in primary schools—will be required solely to replace teachers who leave the profession. Artificial intelligence may be a component of a solution. By supporting two key enablers of teaching: coaching, and assessing, AI-assisted teaching could significantly impact third-world countries and remote locations.
Coaching and assessing require specific skills that are currently beyond the capabilities of machines, such as emotional intelligence, creativity, and communication. Deep learning algorithms could recognize patterns, attitudes toward the learning situation, and affective states using new indicators such as facial expressions, digital interactions, group interactions, and attendance tracking and provide real-time support to students.
In addition to student assessment, AI-powered machines are making progress. Companies such as GradeScope already use computer vision and machine learning to grade students’ work faster than a teacher, starting with deciphering handwriting and remembering the teacher’s initial mark decisions to grade subsequent students automatically. Only work with objectively correct answers, such as math problems, and rule-based learning, such as orthography, languages, and historical events, can be evaluated with today’s technology.
