The intersection of artificial intelligence (AI) and education has become a hotbed of innovation, especially since the advent of large models that fuel various educational applications. Recently, a notable surge of interest has emerged within the education sector, marked by multiple firms unveiling AI-driven products that resonate with both users and market demands. While some companies leverage their foundational business capabilities to refine software and hardware offerings, others are ambitiously exploring opportunities on a global scale.

Experts in education, such as Chao-Hui Chu, a researcher at the Chinese Academy of Education Sciences, highlight that embracing these large AI models is not just a fad, but a necessary evolution. However, they emphasize the importance of adhering to human-centered principles as a cornerstone for deploying these technologies. Ensuring that the use of AI in education remains structured and that its integration promotes equity is crucial to its acceptance and efficacy.

As AI is already making its mark in classrooms, several challenges remain. For instance, the existing metrics for assessing the effectiveness of AI in education are still not universally agreed upon. The dual nature of AI as a tool that can either bridge or widen educational gaps poses a dilemma. Students adept at utilizing AI could leave their less tech-savvy peers behind, giving rise to a digital divide that might ultimately hinder the goal of equitable education, a view that point leads experts to advocate for sound policies that could mitigate these risks.

With the “618” shopping festival on the horizon, the landscape of education-related live-streaming sales is bursting with activity. Brands are incorporating various AI and large model products into their offerings as online hosts passionately promote learning machines and digital study tools across platforms. For instance, the Taobao live stream from iFLYTEK showcases their latest AI-enhanced learning devices, while NetEase Youdao promotes innovative tools such as electronic dictionary pens to boost language acquisition.

Recent data indicates an impressive growth in China's educational technology market, with ancillary electronic products, including study aids, exceeding a remarkable 10 billion yuan. This expansion, marked by a steady annual growth rate of about 30%, underscores the ongoing competition in the underlying technologies that empower these smart products. This competition centers around developing robust large models that drive their advanced functionalities.

Industry veterans like Jianrong Shi view large models as essential frameworks that serve as “tools for creating tools” to bolster education, contrasting with direct consumer-facing products. In July of last year, regulatory support for the integration of large models into the educational domain catalyzed a competitive atmosphere among online learning platforms. By September, Zhuyuan would unveil its self-developed Yinhe large model, marking a significant milestone in adaptive technology for education. Subsequently, in October, iFLYTEK revealed its latest iteration of the iFLYTEK Starfire cognitive model.

Meanwhile, an intriguing development arose this May, as Zhejiang Precision Learning announced the impending release of a native AI learning assistant, alongside nearly 200 million yuan in funding from Alibaba. This product combines their proprietary “Flow Knowledge Mirror” model with Alibaba’s expansive model, showcasing the collaborative potential between such tech giants. Yang Renbin, the founder and CEO of Precision Learning, believes that AI-native applications in the educational sector could lead to revolutionary advances in personalized teaching methods, significantly challenging traditional pedagogies.

As Chinese AI models mature, companies are keenly eyeing commercial viability. Yuanli Technology’s Kanyun large model has already been approved, finding integration into various educational software and hardware products under its brand. Companies like Baidu and Alibaba are also introducing innovative learning devices that emphasize the AI-education synergy. Nationwide, educational institutions and government bodies are ramping up the development of large models, focusing primarily on Business-to-Business (B2B) models to bolster institutional capabilities.

From an investment perspective, education institutions possess distinct advantages due to their vast repositories of data, which can be judiciously harnessed for training models. Yet, despite these advantages, marketing costs continue to soar, creating a disparity between educational innovations and investments compared to their tech counterparts. Reports indicate that major players in this field typically exceed investments in the billions of dollars.

Market dynamics are shifting, as recent ratings improvements from financial institutions indicate growing confidence in educational tech firms like TAL Education. With anticipated growth driven by learning device sales, projections suggest significant revenue milestones, emphasizing positive shifts in operational efficiency. Potential volume gains in learning machine sales, previously projected to reach nearly 90,000 units, could further solidify positioning within competitive markets.

Educational products powered by large AI models are poised not only to drive revenue growth but also to enhance user engagement significantly. A prime example of this success is seen with Youdao’s X6 Pro dictionary pen, which experienced remarkable launch success, selling over 40,000 units in its debut. Following the launch of its new learning machines, iFLYTEK announced extraordinary sales growth, reporting increases in gross merchandise value of 136% and 217% month over month.

As large language models like ChatGPT have ignited a wave of investment and technological commitment, the competition in building robust AI applications, particularly in education, has intensified. Notably, industry thought leaders are scrutinizing the commercial prospects of large models. Following the launch of ChatGPT, numerous education-focused publicly traded companies responded with ambitious initiatives, leading to temporary spikes in their stock values.

Investors increasingly view the educational sector as a fertile ground for AI application, offering substantial potential due to its inherent demand for innovative solutions. Two prevalent business models for sectors nearing market entry typically revolve around B2B and B2G formats, with domestic entities demonstrating higher accuracy compared to international counterparts. Market logic suggests a bifurcated approach: targeting the lucrative North American market alongside the populous Southeast Asian region, leveraging a sizable user base.

It's also noteworthy that the venture into international markets includes mechanisms like government investments in educational resources and collaborative platforms. iFLYTEK has recognized this potential, leveraging its expertise in the development of AI applications tailored for education by establishing regional headquarters in Singapore to expand its Southeast Asian market presence.

As a technological catalyst, large models provide an opportunity to drastically reshape educational tools. Experts like Yang Yuanzhu believe that the multi-modal needs of the education sector surpass other fields, heralding large models as pivotal in realizing effective teaching strategies. Looking ahead, Shi Jianrong envisions enhancements in educational product interactivity and adaptability through these models, making learning processes fluid and less reliant on conventional methodologies.

However, caution is warranted. Chu Chao-Hui has raised significant concerns regarding the overreliance on AI in learning environments. While AI can bolster knowledge acquisition, it may inadvertently instigate barriers in social skills and real-world problem-solving, potentially leading to dependence on technology. Hence, regulation and moderation in the deployment of educational AI tools are essential for ensuring their integration augments, rather than detracts from, holistic development in students.