Houston's employers are embedding artificial intelligence into daily operations, and the demand for workers who can use these tools is growing faster than the supply. The question for the city's schools and colleges is not if AI will reshape the labor market-it already has-but whether students will be ready to compete. Right now, the answer is uncertain, and the window to act is narrowing.
Students are already using AI, often without rules
According to the 2025 Higher Education Policy Institute and Kortext Student Generative AI Survey, 88% of college students now use generative AI for coursework, up from 53% one year ago. Yet more than half of those students attend institutions that formally discourage or ban AI use. The mismatch leaves many navigating tools on their own, without guidance on responsible or effective use.
Harmony's guided model delivers growth
Harmony Public Schools was recently named a Texas State Champion in the Presidential AI Challenge for its Human Core Initiative. Burak Yilmaz, the district's director of instruction, said the framework was built on the recognition that unguided AI use is already hurting students.
"If we ignore AI, our students will just figure out how to use it on their own, and that may not be the healthiest approach," Yilmaz said. "Right now, when it's unguided, AI is being used by young people in a way that is essentially to cheat the system."
The approach, a "pedagogical reversal," requires students to start every AI interaction with their own ideas, use the AI to expand those ideas, and then audit the output. Through a partnership with MagicSchool AI, teachers control environments where AI responds with Socratic questions instead of direct answers. Since the initiative launched, Harmony recorded 66% overall student growth across assessments, with special education and emergent bilingual students each reaching 70% growth-outpacing their general education peers. This fall, all eighth graders will take a mandatory AI literacy course, with plans to extend it systemwide.
"Those high-need student groups aren't just keeping up," Yilmaz said. "They are outpacing the general population's growth rate."
Faculty literacy is the first step
At the university level, Amin Alipour, an associate professor of computer science at the University of Houston, said the most urgent need is not new technology but training for instructors. "We need to start training instructors right now," he said. "AI is very vague to a lot of people. They should know what these systems are and what they do. They are not magic."
Alipour said coherent AI policies are impossible without educators who understand how these tools function and where they fail. The University of Houston has given faculty and students access to Google's Gemini platform and hosted seminars on integrating AI into the curriculum. But Alipour said the pace must quicken. The challenge of building faculty AI literacy is a focus across the broader field of AI for Education.
He also cautioned against treating AI fluency as an independent skill. The graduates Houston needs are those who know their subject well enough to identify when AI produces errors. "Subject matter expertise becomes more important with AI because people need to verify AI outputs," he said. "And that requires knowing your field."
The risk of widening gaps
Research already shows that AI literacy divides follow racial and economic lines. Without deliberate effort, Houston's Black graduates are likely to fall further behind. "The people who are more AI-literate, who have access, who have better networks, they can spot opportunities faster," Alipour said. "Others that don't have that background are going to lose out. It's going to widen the already existing gap in the economy."
Yet he also pointed to AI's potential to open doors that were previously closed to people without technical backgrounds or elite resources. "AI can let more people build tools we didn't have the means to build before," he said. "But only if they have access and the skills to use it."
Why this matters for educators
Educators in Houston and beyond must recognize that students are already using AI, often outside any institutional framework. Clear policies and trained teachers are prerequisites to turning AI from a cheating shortcut into a learning tool that can accelerate growth. The Harmony model shows that when schools guide AI use directly, even historically underserved student groups can post growth rates above the general population. For teachers and administrators, the immediate task is twofold: build your own AI literacy and push for coherent rules that prepare students for a labor market that is already rewarding those who can use these tools well.
Your membership also unlocks:
