Core Summary
This summer, the education and training industry is witnessing a significant shift in funding: money that was previously spent on summer classes is now flowing towards AI learning devices (also known as “AI tutors”). Although the sales volume of learning devices has declined, high-end AI models have become the mainstream, with leading brands forming a monopoly. Each company is losing money in order to capture users, betting on their future influence once AI becomes more widespread. However, AI still cannot replace human companionship and the sense of responsibility that parents seek—this remains the last barrier for the education and training industry.
Detailed Analysis
1. Summer Classes Losing Popularity, AI Tutors Competing for Parents
In the past, parents who signed up for summer classes and those who bought learning devices were two separate groups: the former wanted someone to supervise their children’s learning, while the latter just needed a supplementary tool. But now things have changed. A senior executive from Xueersi (a well-known education company) stated that there is a clear overlap between these two groups of parents. In simple terms, parents are starting to use AI tutors as an alternative to summer classes—no need to send their children to institutions; they can use the devices anytime and accurately identify areas for improvement. It’s similar to how people used to dine out but now prefer to buy high-quality prepared meals to cook at home, with the money flowing from restaurants to the manufacturers of these products.
2. Learning Device Market: Fewer Sales, But Higher Profits?
The data from this year’s 618 shopping festival was somewhat counterintuitive: online sales dropped by 18.2% (to only 360,000 units), but total sales only decreased by 11.4%, and the average price increased by 8.3% to 3,635 yuan. The reason is simple: no one wants low-end learning devices anymore; everyone prefers high-end models with AI capabilities. The proportion of AI-enabled devices has risen from 31% last year to 67% (7 out of every 10 units). The market is becoming increasingly concentrated, with the top four brands (Zuoyebang, Xueersi, iFlytek, and Xiaoyuan) accounting for 86.5% of the market share, leaving smaller brands virtually out of the competition. This is similar to the smartphone market, where knockoff devices have disappeared, leaving only companies like Apple and Huawei selling high-end products—fewer units sold, but higher profits per unit.
3. AI Tutors Are Not Just a “New Packaging for Old Ideas”; Is There Really a Difference?
Traditional learning devices consisted of electronic question banks and recorded lessons that provided answers without explaining why mistakes were made or offered no interactive experience. Modern AI tutors, on the other hand, offer proactive services:
- iFlytek’s devices diagnose students’ learning issues in 3–5 minutes, indicating whether the problem lies in a lack of understanding of concepts or careless calculations;
- Xiaoyuan’s devices use billions of data points to assess knowledge levels (weak/qualified/excellent/proficient), accurately pinpointing areas for improvement.
The key difference is that traditional devices required users to seek information; AI tutors actively identify and address their weaknesses. However, there are challenges: can AI detect when a child is distracted? Can it motivate a child who doesn’t want to study? Many parents find that after using these devices for three months, there isn’t much difference from using older models—the benefits have not fully materialized.
4. Leading Brands Each with Their Own Strategies
The top five brands (Zuoyebang, Xueersi, iFlytek, Xiaoyuan, and Buchong) each have unique approaches to gaining market share:
- Zuoyebang: Focuses on traffic generation. They use a massive question bank to accumulate data and collaborate with Lenovo to create AI study spaces, aiming to expand their user base without focusing on profits from individual devices;
- iFlytek: Emphasizes technology. They utilize advanced models and partnerships with 60,000 public schools to provide detailed analysis of test mistakes, claiming to understand exams best;
- Xiaoyuan: Highlights the effectiveness of their products, claiming that their data-driven assessment model is 98.9% accurate in predicting exam scores, promoting the idea that “mastering the device means excelling in school”;
- Buchong: Expands through physical stores, allowing parents to experience the devices before making a purchase, leveraging the trust associated with established brands;
- Xueersi: Combines various strategies. They rely on their existing reputation in education and offer high-end models (the T6 model costs up to 11,999 yuan), integrating them with their own course resources.
5. Losing Money but Betting on the Future (2027)
These brands are currently losing money due to high hardware costs (screens, chips, handwriting sensors), and the high cost of investing in live streaming (ROI=1). But why continue? They are betting on 2027: by then, the penetration rate of AI in households could exceed 70%, with 190 million AI devices in use. Local chips will make AI more efficient and powerful. The losses now are investments in gaining consumer perception—making parents believe that AI tutors are reliable. Once technology matures, they expect to reap profits, much like how ride-hailing companies initially invested heavily to gain market share before becoming monopolists.
6. Can AI Replace Humans? The Last Barrier Is Human Touch
AI tutors can explain problems, correct assignments, and identify weaknesses, but they cannot provide the same sense of human responsibility that parents seek. For example, they cannot encourage a child when they don’t want to study or bring them back from distraction. Parents pay for more than just tools; they want someone to take care of their children. This is what AI cannot replace, and it remains the last barrier for the education and training industry. The brand that successfully combines AI with human interaction will be the one to survive in the next five years.
Final Conclusion
The money in the education and training market has not disappeared; it has just shifted from summer classes to AI tutors. However, for AI to truly replace humans, it still needs to develop a level of emotional connection that currently lacks.
(End of analysis)