AI-enabled tech is the key to delivering personalized learning

In order to turn learning loss into learning gain, we need bots that serve as personal tutors.

From New Jersey to Texas to California, the pandemic has left our public schools reeling. In addition to lives lost, disengaged students, and overall burnout, schools are facing these challenges:

  • A significant teacher shortage. According to Learning Counsel Research, we’re down by 638,000 teachers–and we already had a shortage before the pandemic.
  • A rise in mental health issues. In the 2022 Crisis Communication and Safety in Education Survey, student mental health ranked number 1 among K-12 education professionals’ concerns (number 2 was faculty and staff mental health).
  • A steady decline of students to private schools, homeschooling, or dropping out. Since the pandemic began, public schools have lost more than 1.2 million students. New York City and the state of Michigan have each lost 50,000 students.

What about learning loss?

Ramesh Balan

Perhaps the biggest challenge our schools face today is learning loss. Texas Education Agency Commissioner Mike Morath says Texas lost a decade’s worth of progress in the last two years, with more than a million children below grade level, exhibiting the state’s largest decline in student knowledge in math and literacy.

The rest of the country is suffering too. Earlier this year, Brookings reported that math and reading test scores from 5.4 million students in grades 3 through 8 between the fall of 2019 and the fall of 2021 fell significantly, with the gaps between low- and high-poverty elementary students increasing by 20 percent in math and 15 percent in reading.

Severe teacher shortages, limited budgets, exhaustion, and the prohibitive cost of 1:1 tutoring means districts are struggling to meet students’ specific learning needs. Teachers lack a dynamic learning delivery solution–not to mention the training–to provide students with high-quality, dynamic learning resources and after-hours support that can free up time for core instruction.

It’s time for outside-the-box thinking to meet every student’s personalized learning needs.

But first, a quick look at why we haven’t scaled personalized learning yet.

Why aren’t all schools delivering personalized learning?

For starters, the technology isn’t smart enough. The offerings are lifeless, with stagnant content and limited (at best) interactivity. Video is better than paper, but it’s only one-way. Tech-based solutions like Quizlet, Kahoot, and Nearpod are great at what they do, but they don’t go far enough.

Ericka Johnson-Allen

In addition to these limitations, today’s technology requires human intervention. Our current digital systems lack the level of intelligence and sophistication needed to deliver a truly personalized experience.

We need to develop bots that will be useful for teachers and students at scale. This process has already started, with the Texas Education Agency approving Amira Learning’s voice A.I.-powered reading tutor as a state-approved solution to address COVID-19 learning loss. The software uses speech recognition to assess mastery and deliver personalized guided practice. Tutors can use Amira’s software for assessment, progress monitoring, and other tasks to try and close skills gaps.

Why not a math, social studies, or science bot? An AP physics bot?

As school districts and state agencies realize the value of having these bots work with children to provide personalized learning support and bridge the gaps–while not overworking our teachers any further–we can begin to move toward desired outcomes while spending much less than hiring an army of tutors.

A vision for a next-generation personalized learning system

We envision bringing to life bots that can handle active learning and understand the district’s subject matter, lessons, and activities. It must recognize the context of what is being taught and the stages by which learning develops so that it can see where a student is deficient. Data is the key to this process. We can train the bot with structured information so that it learns to respond to a student asking a question, taking an assessment, provide contextualized feedback, and so on. In the future, the bot can also learn how to respond to emotional cues, such as offering words of encouragement, thereby addressing social-emotional learning concerns.

Our company, Knomadix, has developed a learning platform that combines artificial intelligence and active learning to deliver personalized instruction and intervention to help maximize student potential. The platform allows state- and district-level curriculum developers, instructional coaches, and teachers to transform static lessons into delivering machine-assisted one-on-one learning support. It also enables publishers and course developers to embed LessonBots into their existing digital courses.

How does this type of technology change the classroom?

With Knomadix, the teacher can use the data to see exactly where students are having difficulties. This kind of granular data will help teachers remediate, catch students up, and intervene before students have failed.

Imagine having 25 virtual teaching assistants that are collecting real-time data and simultaneously aggregating findings into an easy-to-read dashboard. Virtual bot teaching assistants provide support for students so that teachers can navigate through all the student interactions and see who needs what help, where they need help, and gain deep insights into every student’s learning development.

Wouldn’t that be a great step in the direction of learning gain?

Ramesh Balan is founder, chief architect, and CEO of Knomadix. The child of a K-12 educator, Balan has bootstrapped several enterprises and mobile software companies. 

Dr. Ericka Johnson-Allen currently serves as superintendent for one of the largest charter networks in Texas and is CEO and founder of Imagine That Consulting, where she engages leaders through professional coaching and transformational experiences.

More from DA

 

Most Popular