Anthropic just launched Claude for Teachers — a free AI tool for K-12 teachers in the United States. OpenAI has ChatGPT for Teachers, also free, plus ChatGPT Edu, a discounted enterprise subscription for universities. Google has wrapped its Gemini models around its existing education cloud suite. The world’s three largest AI labs are simultaneously giving their products away to educators.
What is happening? The frontier AI labs — OpenAI, Anthropic, Google — are distributing free or heavily discounted AI tools directly to teachers and students worldwide. This is not philanthropy. Morgan Stanley values the global education market at $6 trillion. In students, the labs see tomorrow’s workers, managers, and decision-makers — a generation that will default to whichever AI assistant they used in school.
🔍 THE BOTTOM LINE
The AI labs are pursuing the classic platform play: give the product away for free to lock in users early, then monetise once switching costs are high enough. Education is the ideal beachhead because the users are young, the adoption window is years, and the institutional relationships built now will be worth billions when these students enter the workforce. The $6 trillion market is the prize. The free tools are the entry strategy.
The Tools on Offer
Anthropic’s Claude for Teachers gives K-12 teachers free access to its Claude Cowork AI agent for lesson planning, assignment creation, and automated student reporting. The company also offers Claude for Education at the university level, with a learning mode that asks “how would you approach this problem?” rather than providing direct answers — a Socratic approach designed to guide critical thinking.
OpenAI’s portfolio is broader. ChatGPT for Teachers is a free, self-serve offer for verified US K-12 teachers. ChatGPT Edu is a discounted enterprise subscription aimed at universities. The company’s Education for Countries programme works with governments to bring its technology into national school systems — Estonia already has ChatGPT Edu in the hands of over 30,000 students, teachers, and researchers. OpenAI is running similar partnerships in Greece, Jordan, Kazakhstan, Slovakia, Trinidad and Tobago, the UAE, and Italy.
Google has the deepest existing institutional footprint. Its cloud services already run much of the world’s university infrastructure, and Gemini-powered tools are being layered on top, with more powerful features reserved for paid plans.
Why Education, Why Now
Leah Belsky, vice-president of education at OpenAI and former chief revenue officer at Coursera, frames the strategy candidly: “Every student today is going to grow up with AI. It’s our job to help shape how students use the technology.”
That framing — shaping how students use AI — is doing a lot of work. The company that shapes how a generation learns to interact with AI is the company that owns that generation’s default AI habit. This is the same logic that made Microsoft Word ubiquitous: once you learn it in school, you never unlearn it.
Drew Bent, Anthropic’s education lead, emphasises working “with the educators who really understand the students best.” But the business logic is unmistakable. As Simon Buckingham Shum, professor of learning informatics at the University of Technology Sydney, says: “Let’s not be naive, AI is absolutely a marketplace for these companies.”
The EdTech Response: Partner or Perish
The existing education technology sector — Coursera, Instructure, and dozens of smaller players — has chosen to partner rather than fight. Coursera has integrated its content into ChatGPT. Instructure, which sells learning management systems, is a partner with both OpenAI and Anthropic.
Michael Feldstein, chief strategy officer at the education consortium 1EdTech, describes the dynamic: edtechs are “nervous [and] afraid that the labs are going to [make them] obsolete.” But they retain one advantage the labs lack: data on patterns of usage in education. The labs have the models; the edtechs have the context.
The existential question for edtech is whether the labs need them at all. If Claude can plan a lesson and grade an assignment, the learning management system becomes a thin wrapper around the AI model.
Adoption Is Slow — But the Lock-In Is Long
For all the activity, institutional adoption remains low. A study by the Digital Education Council found that more than 40 per cent of higher education students said AI had not been integrated into any of their courses. Of those whose courses had made use of AI, 42 per cent said it had been only “somewhat helpful,” and 24 per cent reported “limited learning benefit.”
The labs are not discouraged. The free tools are not optimised for today’s adoption — they are optimised for the switching cost that accumulates over years. Estonia’s 30,000 ChatGPT Edu users are not paying OpenAI today. They will.
Feldstein raises the longer-term commercial question: “There’s an open question as to whether schools are going to want to pay out for top chatbots when adequate models will run on their device for free.” The rise of open-weight models — AI whose parameters are released publicly and can be fine-tuned for specific uses — could undercut the labs’ monetisation strategy before it matures.
The NZ Angle
New Zealand’s education system has no equivalent of Estonia’s national AI partnership. Individual universities and secondary schools are making ad-hoc decisions about which tools to adopt, with no central framework. The Ministry of Education’s AI guidance remains advisory, not operational. This means the labs’ free offerings are entering an unregulated vacuum — whatever a school decides to adopt becomes its students’ default AI experience. For a country with no sovereign AI champion, the question is whether NZ education ends up shaped by American AI companies’ commercial priorities rather than by local pedagogical values.
The University of Technology Sydney’s approach — instructors building their own AI products with internal quality assurance — offers a model. Buckingham Shum: “The universities that are doing their jobs are putting design tools in the hands of educators.” NZ institutions could do the same with open-weight models, keeping control of the educational experience local.
❓ FAQ
Are the free AI education tools actually free? Yes, for now. ChatGPT for Teachers and Claude for Teachers are free for verified K-12 educators. ChatGPT Edu is discounted for universities. The business model is long-term lock-in, not short-term revenue. The question is what happens when the free period ends.
Which countries have national AI-in-education partnerships with OpenAI? Estonia, Greece, Jordan, Kazakhstan, Slovakia, Trinidad and Tobago, the UAE, and Italy. Each has a customised version of ChatGPT tailored to local pedagogical style.
Can open-weight models undercut the labs’ education strategy? Potentially. Free, locally-run AI models that can be fine-tuned for specific educational uses remove the switching cost the labs are banking on. But they require technical capacity that most schools do not have today.
Is this good for students? The Socratic approach both OpenAI and Anthropic espouse — guiding critical thinking rather than providing answers — is pedagogically sound. The risk is that the tool that shapes a student’s AI habits is also the tool that captures their future spending.
🔍 THE BOTTOM LINE
The $6 trillion education market is not being conquered by force. It is being conquered by generosity. Free tools for teachers, discounted plans for universities, government partnerships in eight countries — every move is designed to make AI from OpenAI, Anthropic, or Google the default for a generation of students who will carry that default into their working lives. The pedagogy may be sound. The strategy is certainly sound. Whether education institutions are aware of the trade they are making is the open question.
