Is AI the future of test prep?

This article is part of Upstarta series about companies leveraging new science and technology to solve challenges in their industry.

Ever since Socrates taught Plato and Plato taught Aristotle, mankind has known that the best education is given one-on-one by an experienced educator. But that is expensive, labour-intensive and difficult to scale. The result is the imperfect classroom instruction we live with today: large classes, overworked and overworked teachers, a shortage of resources. Teachers spend what little time they have giving personal attention to the best and brightest or at the bottom of the class. The broad middle is often left to its own devices.

Teachers may have a new tool, AI, to address these issues. Innovative forms of the technology, based on computer code that mimics the networks of neurons in the human brain, can discover patterns in how students perform and help teachers adjust their strategies accordingly. “AI teachers” — software systems that students use to interact online — promise to give each student individualized attention, potentially recreating education as we know it.

One of the few companies leading this transformation is Riiiid (pronounced “rid”), a start-up founded in Korea by YJ Jang, a graduate of the Haas School of Business at the University of California, Berkeley. Riiiid already has a strong presence in the Asian test prep app market for the Test of English for International Communication, or TOEIC, which measures English proficiency for business. Now Riiiid is about to enter the SAT and ACT prep market in the United States.

“Education is a complex field closely related to cognition, motivation, interaction with peers, etc.,” wrote Mr. Jang in an email. “We draw insights from learning science, cognitive biology, data science and other related research areas for an iterative experimentation process that is challenging and time consuming – which is why there are only a few players in the market.”

The first computer-guided systems appeared in the 1960s, presenting material in short segments, asking students questions and providing immediate feedback on answers. Because these systems were expensive and computers were far from ubiquitous, research institutes were the main beneficiaries.

In the 1970s and 1980s, systems began to use rule-based artificial intelligence and cognitive theory. These approaches walked students through each step of a problem and hinted at expert knowledge bases. But rules-based systems failed because they weren’t scalable – and programming extensive domain expertise was expensive and tedious.

Mr. Jang was evaluating such systems at Berkeley when a friend, Kangwook Lee, now a professor at the University of Wisconsin-Madison, introduced him to deep learning, a much more powerful form of AI in which algorithms learn on their own using mountains of data. Mr. Jang saw that deep learning could apply to teaching, with systems learning content and student behavior over time.

He returned to Korea and founded Riiiid in 2014, where he and a team of data scientists developed a suite of AI algorithms that track student performance, predict scores, and anticipate when students are losing interest and about to drop out. to crochet. The company has published articles about this work at some of the world’s leading machine learning conferences.

To validate its technology and collect the data needed to refine its algorithms, Riiiid launched a TOEIC test prep app called Santa (Santa, of course, collects data on children around the world). It quickly became one of the best-selling educational apps in Japan and Korea.

Through the app, Riiiid collected data on student interactions, building what is today one of the world’s largest public education datasets, called EdNet. But Riiiid has struggled to collect enough data to generalize its AI system to the wider education field.

“It is difficult to collect AI-trainable, multimodal data in diverse learning environments,” Mr. Jang.

For now, the company is focusing on the $300 billion test prep market, where data is easier to collect, and is partnering with education companies in different parts of the world to develop test prep apps. Earlier this year, Riiiid partnered with Casa Grande to launch an app called OE Sabre, which helps students in Colombia prepare for the entrance exam for the country’s Saber 11 university.

Riiiid’s success attracted a $175 million investment from venture capital giant SoftBank’s Vision Fund II, bringing the company’s funding to approximately $250 million.

Now, Riiiid introduces an AI-powered preparation platform for the SAT and ACT college entrance exams. The product, R.test, due for release in January (pricing to be announced), predicts standardized exam scores in a quarter of the time it takes to complete a full mock test. By answering 30 questions, students receive an analysis of their weaknesses and advice on how to improve, including an AI-curated selection of relevant practice questions. Riiiid says the idea is for students to practice with the app and take the actual exam with some confidence in what their final scores will be.

“I really liked this because we can just use it at home instead of hiring a tutor,” says Esther Yi, a parent in Georgia who tested an early version of the platform. She found the R.test analysis particularly powerful. “My 10th grader will definitely benefit from this,” she said.

Oscar Torres, a high school math teacher in Chicago who has been trying Riiiid’s system, said he liked R.test because it assesses students’ knowledge in real time without relying on previous test scores. “As AI develops, I can see it becoming a better and stronger resource for us,” he said. “As teachers, we need to solve problems in real time, and AI can help us tremendously.”

But the company’s goal remains broader than test preparation. Mr. Jang said R.test is part of an effort to collect data and prove the effectiveness of its algorithms in other domains. Riiid researchers continue to develop new architectures and higher-performing AI models that can monitor student behavior, track student knowledge, and select the best content to study at any given time.

“Our algorithms can predict student test scores with astonishing accuracy, a moving number that serves as a kind of square root,” he said. “The more students follow the algorithmic recommendations, the higher their predicted score will rise.”

Mr. Jang believes that soon teaching will no longer be based on guesswork or intuition, but on data. And that’s perhaps the company’s biggest challenge: Collecting that data, he added, is the sticking point, as privacy concerns make data collection in schools a complicated issue. (Riiid says its apps don’t collect personally identifiable information from users.)

To address these concerns, Riiiid helped create the EdSAFE AI Alliance, a global, cross-industry alliance of companies, nonprofits and education technology associations to develop benchmarks and standards to promote the safe and responsible use of AI in education. guarantees.

“The dream,” said Mr. Jang, “is to integrate these algorithms into a comprehensive system that can teach any subject to anyone, anywhere.”