EduNational is a New York–based company that develops artificial intelligence programs to evaluate and treat language-related neurological conditions, with a particular focus on dyslexia and other learning disabilities. Pulse 2.0 interviewed EduNational CEO Coral PS Hoh, PhD, to gain a deeper understanding of the company.
Dr. Hoh’s Background
Could you tell me more about your background? Dr. Hoh said:
“My undergraduate and graduate education is in Linguistics, the scientific study of natural language. As a cognitive science, Linguistics uses people’s verbal behaviors and language outputs to make inferences about brain functions and processes. I developed a method for studying brain processing in this functional, non-invasive way through my BA and PhD theses, followed by 30 years of fieldwork in clinical linguistics. This method, in its various applications, led to peer-reviewed publications and referee work.”
“This linguistic method motivated the creation of my current venture in AI. I collaborated with mathematicians and software engineers throughout my career. I translate linguistic models into mathematical ones for coding.”
“There’s also the entrepreneurial side of me. All my siblings are successful entrepreneurs. Through their many enterprises, I learned about starting and managing a business.”
Formation Of The Company
How did the idea for the company come together? Dr. Hoh shared:
“I founded EduNational LLC in 2007 early in my career to apply my doctoral training. At first, I worked with children who had special academic needs. I kept encountering cases with dyslexia as this reading/learning disability affects 1 in 5 people.”
“As I got to know the children with this condition, their struggles at school, the challenges faced by their families, and the whole dyslexia landscape, I realized that a scalable solution was the only answer.”
“Without scalable technology, dyslexia specialists work one-on-one or in small groups in special education. This leaves out about half of the 10 million children with dyslexia in the U.S. Moreover, students who are receiving the specialist intervention still generally cannot read on grade level. Dyslexia is the largest category in special education, which collectively costs U.S. schools over $120 billion a year.”
“EduNational evolved into a tech startup as our team focused on building an expert system to screen and correct reading difficulty, or dyslexia. We called the AI system ‘Dysolve,’ blending ‘dyslexia’ and ‘dissolve.’
“I enlisted the help of my family, who have the expertise needed. My daughter, Charisse Haruta, worked with me to design the architecture of this autonomous AI system. Charisse is a graduate of Carnegie Mellon’s School of Computer Science and Cognitive Science program. We were granted patents for the technology in the U.S., Canada, China, Japan and Korea, with others pending.”
“During product development, our team was fortunate to have the counsel of my husband, Evan Haruta, a software engineer known for mitigating and resolving large operating system crashes in critical situations worldwide. Dysolve AI has provided uninterrupted service since beta testing in 2017.”
Favorite Memory
What has been your favorite memory working for the company so far? Dr. Hoh reflected:
“The first user of Dysolve AI was Alex G., a 3rd grader in 2017. None of us interacted or worked with Alex. Dysolve AI handled his evaluation and training autonomously. Four months later, Alex’s dad came to our office and showed us his reading scores from his school’s quarterly standardized assessments. It looked like a checkmark. Since joining Dysolve, Alex’s scores rose sharply from the 20th percentile to the 50th in 4 months.”
“It was a eureka moment for our team. We realized then how fast Dysolve AI could work and why. Since then, many parents had told us about their eureka moments with Dysolve AI. They had gone through years of trying other interventions but none could pinpoint their children’s problem. Dysolve AI found their processing problems right away and focused on clearing them. Their parents were overjoyed: Finally, a program figured out their children’s reading problem.”
“Every time we get a child past their reading difficulty, we celebrate this life-changing moment. We have lots of favorite memories.”
Core Products
What are the company’s core products and features? Dr. Hoh explained:
“Dysolve AI is a subscription-based program for dyslexia and other learning disabilities. It comprises an evaluative and a training component. The former yields an Evaluation Report on the user’s language processing problems. The latter involves a weekly regimen to correct these difficulties, usually in 1-2 years.”
“Dysolve is designed to be as easy to use as possible for young children and up. It is a plug-and-play program. The user merely logs in at dysolve.com to train through games. Parents and teachers have their own dashboards to see evaluation results and progress reports.”
Challenges Faced
What challenges have Dr. Hoh and the team face in building the company? Dr. Hoh acknowledged:
“The quickest way to deliver the solution to children is through the public school system. Our challenge is getting the message to school administrators.
Currently, schools are facing these crises in addition to funding cuts:
- a) ballooning special ed costs – 1 million more students added just in the past 5 years
- b) special ed teacher shortage
- c) deepening reading crisis – more than two-thirds failing post-pandemic
Dysolve AI has these advantages:
- a) cost effective – on a per pupil basis, Dysolve costs less than 10% of current special ed spending on a student throughout school
- b) automated program needing minimal supervision
- c) efficacious
Dyslexia has long been thought to be lifelong. Our position that dyslexia is correctable has stood up to challenge. Our approach is a logical one: create a method to locate the source of the problem. If you can find the source, you can correct it.”
Evolution Of The Company’s Technology
How has the company’s technology evolved since launching? Dr. Hoh noted:
“Let me describe this in terms of meeting certain goals at various stages of development. Our first priority was to make sure that the technology works, to resolve language processing issues. Next, we had to build out the infrastructure to accommodate the diversity of language processing problems across the population. Then we completed the user interface for individual subscribers in 2020 and for schools in 2023.
Now we are focusing on improving user experience, especially given our unique value proposition of changing language processing in the brain. Our next release this year will feature enhancements to the game experience.”
Significant Milestones
What have been some of the company’s most significant milestones? Dr. Hoh cited:
“All the firsts.
2016
– First patent for a generative method in the diagnosis and treatment of language-related disorders
2017
– First AI system to correct a neurological disorder in an individual autonomously without any human input
2024
– First autonomous AI system to advance a student from the 1st percentile to the 50th in state reading assessment in 3 months
2022-2025
– First clinical trial shows positive effect in preliminary results (n = 750+). This is the first such large-scale randomized controlled trial on the effect of an online intervention on the lowest-performing readers in grades 3+. Study results will be released in October.”
Customer Success Stories
When asking Dr. Hoh about customer success stories, she highlighted:
“There are many. We documented some of the early ones in our book Dyslexia Dissolved: Successful Cases of Learning Disabilities, ADHD and Language Disorders. Many children with dyslexia have co-occurring conditions. One 5th grader who used Dysolve AI had dyslexia, autism, ADHD and ODD (Oppositional Defiant Disorder). He could barely function at school and had meltdowns often at home. After Dysolve, he was getting 90s in high school. Today, he is a well-adjusted young man making plans for college.”
“Over and over again, Dysolve AI has shown this ability to turn some of the hardest cases into successes. The successes often involve students after 3rd grade, when other interventions lose their effect, when special ed teachers run out of things in their toolbox to try.”
Funding/Revenue
When asking Dr. Hoh about the company’s funding and revenue details, she revealed:
“Right from the start, paid subscriptions subsidized product development. External funding was not needed. Now that the product is in commercial use, we are seeking $3 million to expand first in New York, our base. The venture is projected to be profitable in the first fiscal year after this capital raise.”
Total Addressable Market
What total addressable market (TAM) size is the company pursuing? Dr. Hoh assessed:
“The New York TAM for children is $480 million. The U.S. TAM is $10 billion. U.S. schools collectively spend over $120 billion a year on special ed, with dyslexia representing the largest share of students (~32%). There is also a market for adults with dyslexia, largely untapped.”
Differentiation From The Competition
What differentiates the company from its competition? Dr. Hoh affirmed:
“We have the first and only technology to create a game in real time for a specific user during a session, analyze the responses and use the results to generate the next game in realtime, and so on continuously. This capability to generate a new game specifically for each person is unique to Dysolve AI and is patented.”
“This generative capability enables Dysolve AI to be highly sensitive to a person’s processing difficulties so as to correct them. Indeed, Dysolve AI has resolved reading difficulty in individuals since 2017. Last year, a dozen users reported in the press the end of their reading difficulty. No other method has succeeded in correcting reading difficulty in older students.”
“Dysolve AI is the only program linking the screening/evaluation with intervention so that the former provides the blueprint for the latter. With other dyslexia or reading programs, screening is separate from intervention.”
Future Company Goals
What are some of the company’s future company goals? Dr. Hoh emphasized:
“We plan to release Dysolve AI in other languages abroad.”
Additional Thoughts
Any other topics you would like to discuss? Dr. Hoh concluded:
“A solution like Dysolve AI is urgently needed now because of the pandemic’s effect on language development. Studies are beginning to show significant developmental delays in the group who were 0-6 years old during COVID lockdowns. Clinics and schools are reporting a doubling of the incidence of language and communication problems in some regions.”
“Given Dysolve AI’s technological advantage, it can turn things around quickly. Adoption carries little risk as there are minimal indirect costs, no integration with the school curriculum or IT system. It has been tested rigorously for the past 8 years, with successful cases retaining their reading and academic gains.”
