California Community Colleges Spend $500K Yearly on AI Chatbots That Can’t Even Name Their Own President

• Three California community colleges spending up to $500K annually on AI student assistants
• Chatbots handle general questions well but fail on specific campus-related inquiries
• East Los Angeles College’s bot couldn’t correctly identify the college’s own president
• CalMatters investigation reveals major gaps in AI implementation for education
• Malaysian universities should take note before investing in similar technologies

The $500K Question: Are AI Chatbots Worth It for Student Support?

California’s community college system just learned an expensive lesson about AI implementation. According to a CalMatters investigation, three California community colleges are collectively spending up to $500,000 per year on AI chatbots designed to help students with financial aid, admissions, and other critical services. The problem? These digital assistants are often failing at the very tasks they’re meant to automate.

The chatbots, implemented as part of a pilot program across the state’s community college system, were marketed as revolutionary tools that would streamline student services and reduce administrative burden. Instead, they’re leaving students frustrated and administrators questioning the return on investment.

When General Knowledge Isn’t Enough

The core issue isn’t that these AI chatbots are completely useless—they can handle general questions about application processes, financial aid deadlines, and basic enrollment procedures. However, when students ask specific questions about their own campus, things go downhill fast.

Perhaps the most embarrassing example comes from East Los Angeles College, where the deployed AI chatbot couldn’t even correctly name the institution’s own president. This fundamental failure raises serious questions about how these systems were tested before deployment and whether they were properly trained on institution-specific data.

Students seeking help with specific financial aid issues, campus-specific scholarship opportunities, or detailed transfer requirements often find themselves hitting a digital wall. The chatbot either provides generic information that doesn’t apply to their situation or simply admits it doesn’t know the answer.

The Implementation Gap

This situation highlights a critical challenge in educational technology: the gap between AI capabilities in controlled environments and real-world deployment. Large language models can excel at answering general knowledge questions and passing sophisticated exams, but they struggle with the granular, institution-specific information that students actually need.

Community colleges serve some of the most vulnerable students in American higher education—many of whom are first-generation college students working part-time jobs and navigating complex financial aid systems. When an AI chatbot fails them, the impact extends beyond mere inconvenience into potentially derailing their educational aspirations.

The California community college system isn’t alone in this struggle. Educational institutions across the country have been rushing to implement AI solutions, often without adequate testing or consideration of the specific challenges facing higher education.

What This Means for Malaysia

As Malaysia continues its own digital education initiatives, this California cautionary tale offers valuable lessons. Malaysian universities and colleges are increasingly exploring AI-powered student services, from chatbots handling enrollment inquiries to AI tutors assisting with coursework.

The key takeaway is straightforward: don’t spend big without proper testing. Before investing hundreds of thousands of ringgit in AI solutions, institutions should conduct thorough pilot programs with real students, specifically testing the system’s ability to answer institution-specific questions.

Malaysian universities should consider starting small—implementing AI assistants in limited capacities and measuring their effectiveness before scaling up. The country’s existing focus on digital education through initiatives like the Malaysian Digital Economy Corporation (MDEC) programs provides a strong foundation, but that foundation should include rigorous testing protocols.

Local institutions can learn from this California experience by demanding concrete proof of AI performance on campus-specific queries before signing any contracts. Generic demonstrations mean nothing when a student asks about their specific scholarship eligibility or transfer requirements.

Our Take

This is yet another example of technology being deployed before it’s ready for real-world use. The $500K spent by California community colleges could have been used to hire more human counselors, improve existing student services, or fund more carefully designed pilot programs.

AI has tremendous potential in education, but institutions need to stop treating it as a magic solution to staffing challenges. The most successful implementations will likely be hybrid models—AI handling routine inquiries while human staff focuses on complex student needs that require empathy, judgment, and institution-specific knowledge.

For Malaysia’s educational institutions, the message is clear: be skeptical of vendors promising revolutionary AI solutions, test thoroughly with real use cases, and prioritize student outcomes over technological novelty.

Source:

California colleges spend millions on faulty AI systems: ‘The chatbot is outdated’