Equity, Evidence and Empowerment for EdTech

Edtech is, unfortunately, dominated by more hype than evidence. Asyia Kazmi, OBE (Global Education Policy Lead at the Gates Foundation), Shaveta Sharma-Kukreja (CEO and MD of the Central Square Foundation), and Janhvi Maheshwari-Kanoria (Executive Director of Innovation Development at Education Above All) were brought together by Anjali Nambiar of the International Education Funders Group (IEFG) to discuss this at this year’s CIES conference.

We echoed a simple and surprisingly unconventional message:

There is a need for EdTech to be impactful, but we are not there yet. In order for it to work for the learners who need it the most, we need EdTech to be grounded in the realities of these learners and their learning environment, backed by evidence, and built to integrate with the system.

Our convictions converged on three fundamental principles: Equity, Empowerment and Evidence.

We understand that education is fundamentally about human connections, but technology could help us leapfrog some of the inherent gaps between the haves and have-nots. As artificial intelligence accelerates technological advancement, a worrying gap threatens to widen between high-income countries that can afford cutting-edge solutions and low- and middle-income countries (LMICs) that risk being left behind. This digital divide isn’t just about access to technology; it’s about how it’s used and who gets to shape the future of learning. Therefore, local, context-driven, informed innovation solving real problems isn’t just preferable; it’s essential.

There are multiple deep-rooted challenges facing the global education landscape, including:

1. Low literacy levels: By the age of 10 years, 9/10 children in low-income countries cannot read proficiently (Source: UNESCO, World Bank), while 9/10 children in high-income countries can.
2. High out-of-school population: 251 million children and youth are out of school (Source: UNESCO Global Education Monitoring Report, 2024).
3. Poor teacher supply: We are predicted to have a global shortage of 44 million teachers by 2030 (Source: UNESCO Global Report on Teachers).
4. Children in conflict: There are 473 million children affected by conflict (Source: UNICEF) that are in need of trauma-informed education, and 47 million displaced and refugee children with no portable and verifiable credentials (Source: UNHCR).
5. Digital divide: There are 2.9 billion people around the globe not connected to the internet (Source: ITU) and in low-income countries, 90% of adolescent girls and young women aged 15-24 are offline, compared to 78% of adolescent boys and young men of the same age who do not use the internet (Source: UNICEF).

Tech innovation needs to be designed with these contexts and needs in mind. The question is not whether to have EdTech, it’s already here and will be even more prevalent with AI. The question is how best to have it for underserved populations. We have an opportunity to learn from our past mistakes. The first question we need to ask ourselves is: What is the problem that we are trying to solve, and can technology play an impactful and meaningful role in the solution?

Led by this thinking, Education Above All’s Digital School project creates an alternate accredited hybrid learning solution for the out-of-school adolescents in Zambia to support them using bespoke, contextualised digital content and personalised adaptive learning and facilitated teaching to be enrolled in the EMIS system and sit for the Grade 7 exams.

The word adapting is common in ed-speak, as we usually are working to make some centrally developed curriculum more appropriate for specific user groups. In the past, we struggled to translate, change formats and bring local examples to curriculum content – but technology gives us the opportunity to have cost-effective ways of ensuring user-centred design principles to ensure local appropriateness and relevance.

The second design question to be asked is: Is this focused on our users, and will it be adopted?

With 90% of Indian students enrolled in government or low-cost private schools, most families cannot afford commercial EdTech platforms. The shortage of quality educational content in local Indian languages, which are the primary medium of instruction for most students, further limits access to digital learning resources. Developing public educational resources is therefore essential to provide affordable, vernacular digital content that can reach students at scale. This is demonstrated through Central Square Foundation’s Tic Tac Learn, which has over 14,000 ‘bite-sized’ videos in six local languages covering maths, science and social studies. The suitability of the format, language and need of the students drove the adoption to 350 million views on YouTube.

Finally, designing for long-term sustainability in low-resourced contexts. We need to ask ourselves: What digital infrastructure and what digital competencies exist? Otherwise, we would be working on designing interventions that require constant external funding for hardware, high-speed connectivity, electricity or training for digital literates.

The Gates Foundation is supporting platform organisations that seek to build from these starting points, tailoring solutions to local contexts. This is seen in the work of the AI taskforce and the work of organisations like Central Square Foundation and Ai-for-education.org.

Generative AI is accessible, by text or by voice, and in natural language. It could allow us to democratise technology.

Thinking from a systems lens, to avoid duplication and support coordination, we should start with a common taxonomy. Effective communication across the EdTech ecosystem requires more than good intentions; it demands shared frameworks and a common language that can enable comparability, bridge different contexts and stakeholder groups. The development of taxonomies, evaluation standards, and impact measurement tools represents critical infrastructure for the field’s maturation. These frameworks can serve multiple purposes: help funders make more informed decisions, enable organisations to learn from each other’s experiences, and provide educators and policymakers with tools to evaluate and compare potential solutions. Most importantly, they will create accountability mechanisms that can drive the field toward greater focus on equity-informed learning outcomes rather than technology adoption focused merely on usage. The Gates Foundation is supporting public goods like quality benchmarks and evaluation standards, a shared language that helps countries and developers make sense of what solutions are out there, and which ones are worth scaling.

In India and across the Global South, EdTech appears to be booming. But amidst all the scale and funding, there’s a quiet crisis: very little is known about what actually works to improve learning. “Everyone’s focused on scaling access. But what about impact on learning outcomes?” says Shaveta, CEO and MD of Central Square Foundation. “What about the parents, schools, and governments who need tools to make smart, evidence-based decisions?”

To fill this gap, CSF built EdTech Tulna, an independent rating framework that enables policymakers and users to assess the quality and effectiveness of EdTech products. It’s one of the only demand-side tools of its kind globally.

This is the kind of work that’s hard to fund, too early-stage for investors, and not exciting for press releases. But it’s the work that enables accountability, and ultimately, helps shift power back to learners and communities. Tulna aims to foster the quality-led and evidence-based adoption of EdTech in the ecosystem by creating EdTech quality standards to define “what good looks like”, building evaluation toolkits and training to support users in evaluations, and publishing reviews of products that drive demand and shape supply.

Also, on the demand side are the students – in a world where 7/10 children in LMICs have smartphones (Source: World Economic Forum’s Global Risks Report 2024), we know that technology and AI are already in their hands. Can we support students to be discerning users of technology? While governments debate the benefits of AI, the media, gaming companies, algorithms, and pernicious chat agents are already at work. In order to prepare our children for a world that is powered by technology, Education Above All is developing DigiWISE – a curriculum designed to help children explore their digital personas and become more critical consumers and producers of content online. The same sentiment is echoed with CSF’s AI Samarth programme, designed to support AI literacy and skills.

“The time for technology is now, it’s also time we simplify and democratise it. It cannot be something distant, complicated and inaccessible, but participatory for communities to create their own learning in ways that make sense to them!” says Janhvi, who leads the innovation work at Education Above All.

EdTech can likely help unlock many of our challenges, but how do we know this? Can we actually prove it works? Without robust evidence, the field continues to cycle through promising solutions that ultimately fail to deliver transformative outcomes that are sustainable and at scale.

Asyia says, “There are so many thousands of EdTech products, and yet only around 7% of products targeted to LMICs have been evaluated for impact on learning, and often these studies are small scale!”

Interestingly, even in more affluent contexts, the main metrics are driven by funders hungry for scale. The result is often a landscape where marketing claims outpace evidence, and adoption or engagement rates are confused with effectiveness measures. (After all, as Asyia asserts, she was very engaged in Candy Crush, but that didn’t help her learn much.)

The additional persistent challenge in EdTech development is the tension between rapid innovation and rigorous evaluation. Research is expensive and time-consuming, particularly in the early stages when organisations are focused on product development and initial scaling. Our aligned message is that just as we need to customise teaching, so too we need to customise research – research needs to be continuous, informing each stage of product development and scaling rather than a one-and-done approach.

“There is no one-size-fits-all with research, let’s focus on #research-at-right-level! In early stages, we look for important metrics, including learning, adoption, engagement and fit for context are being tracked and publicly available. We need to see that the evidence is clear and actionable, so each iteration gets better,” says Janhvi.

The built-in advantage of EdTech, especially student-facing tech, is that it can help collect important and timely data on learning outcomes. In the early phases of development, using internal data is important to build a case around how effective the EdTech product really is, what groups of learners it is benefiting and who it isn’t, how much usage is required in terms of time on the solution, and what needs to evolve and change. It is also important, of course, to have full cost data. Funders can then consider if an initially expensive but impactful solution can become affordable to LMICs through economies of scale and innovative partnerships. EdTech solutions focused on teachers may well fall into this category – we don’t have much evidence yet for this assertion, but one we are exploring.

“It is also important that we funders, as well as adopters of EdTech tools, demand the data and make decisions that are informed by this data. In this effort, we are trying to make data more accessible and public, and generate a body of evidence that improves the salience of EdTech uptake and adoption,” says Shaveta.

The funder world could be accused of being more competitive than collaborative, with multiple competing priorities and perhaps a desire for influence. The three of us found synergy as we nodded in agreement on hearing the same aspirations and shared frustrations.

With scarce funding, we need to reorient to the goals that matter because, despite our best efforts, we continue to risk failing the very children we care about. We need to work together and be critically honest about what we are looking for and what we are missing, so that our collective efforts can be concentrated for more potency.

The future of EdTech must be shaped by evidence, guided by equity principles, and grounded in the realities of learners worldwide. This requires new approaches to how we fund, develop, evaluate, and scale educational technologies.

For funders, this means building partnerships, prioritising real impact over marketing metrics, supporting locally appropriate innovations, and investing in the research infrastructure needed to build a robust evidence base.

• For technology developers, it means engaging with students, teachers, governments and communities as partners rather than customers; making use of evidence to design and implement at each stage; designing for sustainability and accessibility, and committing to making public the evaluation of and impact on learning outcomes.
• For policymakers and educators, it means demanding evidence before adoption, prioritising solutions that address local needs, maintaining healthy scepticism about technology’s transformative potential without proper implementation and support, and being clear-eyed about funding EdTech at scale.

The stakes couldn’t be higher. With hundreds of millions of children lacking access to quality education, combined with rapid developments in technology, we cannot afford to continue the cycle of hype, investment, and disappointment that has characterised too much of the EdTech landscape. The time has come to rebuild EdTech systems that work for the majority of the world’s children, not just the richest few.

We will try and do this together. The conversation has started, but we’re just getting started.

Transparency Notice: Central Square Foundation (CSF) receives funding from the Gates Foundation to support its work in improving the quality of education for students from low-income backgrounds in India’s primary schools. In addition to Gates Foundation support, CSF also receives funding from various other philanthropic donors.