Universities are racing towards AI. [But are their data foundations ready?]

Across the higher education sector, AI has quickly moved from experimentation to strategy. Universities are now launching AI roadmaps, exploring different tools, testing automation tools and evaluating how AI can enhance student services.

Yet our latest research suggests a critical question is being overlooked:

"Are institutions building AI on foundations they can truly trust?"

At binary10, we believe technology should help organisations make better decisions and deliver better outcomes. That's why our latest report on higher education focused not just on AI adoption, but on the quality, governance and accessibility of the data that powers it. we help organisations make data-driven decisions they can trust.

The AI opportunity is real:

There is no doubt that AI is already transforming higher education.

Our research found that 92% of UK students use AI tools daily for activities such as assignments, research and productivity support. Universities are responding by introducing AI initiatives across student services, administration and academic support.

Some institutions have already established dedicated AI and data teams, while others are actively piloting new technologies linked to strategic objectives and institutional KPIs.

For university leaders, the opportunities are compelling:

• Faster responses to student enquiries

• More personalised student support

• Improved operational efficiency

• Better use of staff time

• Enhanced decision-making

  
  

If implemented effectively, AI could help universities deliver more responsive and personalised services while improving the overall student experience.

But there is a problem:

While enthusiasm for AI is growing, readiness is far less consistent.

During interviews with university professionals, we found a striking divide between strategic ambition and operational reality. Some institutions have mature AI strategies; others admitted staff simply do not know what is happening centrally regarding AI preparation.

One interviewee summed it up perfectly:

At the same time, almost every participant recognised the same underlying challenge:

The risk: AI built on unreliable data

This is where many digital strategies begin to unravel.

AI systems learn from the information they are given. If that information is incomplete, inconsistent or fragmented across multiple systems, the outputs become equally unreliable.

As Steve Smales, COO at binary10, explains:

That insight reflects one of the strongest themes emerging from the research: universities recognise the value of AI, but many are still wrestling with fragmented systems, inconsistent data quality and unclear ownership of critical information.

What AI sees when your data is fragmented:

Many universities already struggle with:

• Multiple student record systems

• Duplicate records

• Conflicting information

• Inconsistent reporting

• Lack of a single source of truth

• Legacy systems holding critical information

Now imagine introducing AI into that environment.

Instead of creating efficiencies, universities risk generating:

Incorrect student communications: 

AI-powered communications can only be accurate if the underlying student records are accurate.

Poor student support decisions: 

Incomplete records may result in vulnerable students being overlooked or receiving inappropriate interventions.

Conflicting information:

AI assistants pulling information from multiple disconnected systems can provide different answers to the same question.

Loss of Trust:

When users cannot trust AI outputs, adoption falls and confidence in the wider transformation programme suffers.

For students, these issues are not technical problems. They become frustrations that directly impact the student experience.

Strategic ambition vs operational reality:

One of the most interesting findings from the report wasn't about technology at all.

Many universities now have:

• Data strategies

• AI strategies

• Digital transformation programmes

• Innovation roadmaps

However, the underlying data architecture often remains fragmented across legacy systems and local repositories. Some institutions are piloting AI tools while their integration models and data foundations remain immature.

As James Blake, CEO of binary10, explains:

This is the difference between being enthusiastic about AI and being truly AI-ready.

Three Questions Every CIO, CTO and Head of Data Should Ask

Before investing further in AI programmes, consider:

1. Do we trust our data?

Would decision-makers confidently act on the information available today?

If the answer is "sometimes", there is more work to do.

2. Do we know where critical data lives?

Many institutions believe they know their data landscape, but the reality is often far more complex. Fragmentation remains one of the most common challenges identified in the research.

3. Could we explain an AI decision?

Strong governance, lineage and documentation become increasingly important when AI is involved. Universities need visibility into where information comes from and how decisions are reached.

What good looks like:

Our research suggests that successful AI adoption starts long before the first AI tool is deployed.

The foundations include:

• Build reliable data foundations:

  Create trusted, high-quality datasets that can support strategic decision-making.

  
  

• Improve AI-critical data

  Prioritise the information that will power AI models, copilots and automation services.

  
  

• Focus on Governance

  Establish ownership, data standards, lineage and documentation before scaling AI initiatives.

  
  

• Start with use cases, not technology

  Identify genuine institutional challenges that AI can solve and work backwards to understand the data requirements.

• Invest in Data Literacy

  Ensure staff understand both the opportunities and limitations of AI technologies.

The bottom line:

AI has enormous potential to transform higher education.

But AI readiness is not determined by the sophistication of the technology being implemented. It is determined by the quality, accessibility and trustworthiness of the data beneath it.

At binary10, we see this every day. Technology can accelerate progress, but only when the foundations are strong. Universities that invest in data quality, integration, governance and trusted information today will be the ones that unlock the greatest value from AI tomorrow.

Because ultimately, the goal isn't simply to deploy AI.

The goal is to make better decisions, deliver better student experiences and create outcomes that institutions can trust. That's what making data-driven decisions you can trust really means.

Want to explore this topic in more detail?

This article is based on findings from our 2026 Higher Education Data Challenges Report, where we interviewed data and technology professionals from 15 UK universities to understand the biggest challenges facing the sector today.

The research explores three recurring themes:

• Too many systems, no single source of truth

• Data quality you cannot trust

• AI readiness built on shaky foundations

The report also provides practical recommendations for CIOs, CTOs, Heads of Data and Transformation leaders looking to build stronger data foundations for the future.

Download the full 2026 Higher Education Data Challenges Report to discover the research, insights and recommendations in full. LINK

Article by:

Raina Das

Follow me on LinkedIn

FAQs:

1. What is AI readiness in higher education?

AI readiness in higher education refers to an institution's ability to successfully adopt and scale artificial intelligence technologies. This includes having trusted data, clear governance frameworks, strong data quality, secure infrastructure and staff who understand how AI can be used responsibly. Without these foundations, AI initiatives can struggle to deliver meaningful outcomes.

2. Why is data quality important for AI?

Data quality is critical for AI because artificial intelligence systems rely on data to generate insights, recommendations and decisions. If data is inaccurate, incomplete or inconsistent, AI can produce unreliable results and amplify existing problems. High-quality, trusted data helps organisations make better decisions and improves confidence in AI-driven outcomes.

3. How can universities prepare for AI adoption?

Universities can prepare for AI adoption by focusing on their data foundations first. This includes improving data quality, establishing data governance, creating a clear data strategy, integrating disconnected systems and ensuring key datasets are accurate and accessible. Successful AI programmes are typically built on strong data management practices rather than technology alone.