Is AI Breaking Down Data Silos?

After years of AI hype—and some recent skepticism—organizations are now seeing real value from their AI initiatives.

While some highly publicized reports have claimed that generative AI projects experience high failure rates, a majority of surveyed organizations report positive outcomes. In a recent survey of The Information’s readers, 61% of readers say their organizations are seeing positive value from their AI initiatives.

Much of the early excitement around AI centered on the promise of creation: automated writing, video, coding and design. Many organizations are now focusing on tools that help them connect, summarize and analyze the vast quantities of data they currently store—much of which is trapped in silos and therefore largely goes unused.

More than half of readers say their organization has either implemented AI solutions specifically aimed at breaking down data silos, or plans to do so soon. And the organizations that have deployed AI to break down these silos have seen a range of benefits, including faster decision-making, better visibility into operations and increased collaboration across departments.

Industry leaders such as SAP are integrating AI directly into core business applications — from finance to supply chain and HR — to connect the data that traditionally sits in separate systems. This integrated approach enables organizations to generate more accurate insights and faster recommendations by drawing on semantically rich, real-time data.

In short, readers say AI tools are helping their organizations achieve some of their most important business outcomes.

The Danger of Data Silos

Organizations produce more data than ever, but much of it is hidden in silos, making it hard to maximize value. According to Forrester, knowledge workers spend nearly 30% of their time searching for information and disconnected systems—often due to legacy tech, mergers, or shadow IT—creating barriers to collaboration and insight.

Data silos can lead to a number of business problems, including cumbersome collaboration, incomplete insights and inefficient operations. But most importantly, perhaps, they slow down decision-making, as leaders simply can’t access the information they need to help them understand their options and make the best possible choice.

In The Information survey, 56% either agree or strongly agree that data silos slow down decision-making in their organizations.

By harmonizing data and embedding AI natively into business applications, companies like SAP are reducing the operational costs of fragmented systems,  explains Irfan Khan, chief product officer for SAP Data & Analytics. “For example, SAP’s business data fabric approach ensures that data  is accessible for analytics and AI, preserving business context and reducing silos, making it easier for organizations to move from raw data to real-time business outcomes.”

AI tools have the potential to unify data silos, resulting in faster, more informed decisions, improved collaboration, greater agility and even a lower total cost of ownership through the elimination of redundancies and manual processes. Breaking down data silos can also improve businesses’ ability to take advantage of other AI tools and processes, which thrive on harmonized, trustworthy data troves.

Currently, only 40% of readers say their organizations are currently using AI tools to break down data silos, but another 19% say their organizations plan to do so in the next year. Those that are already using AI to break down these silos are already seeing a number of benefits—and other respondents expect to see similar benefits if their organizations implement targeted AI solutions.

One industry example of this is SAP Business Data Cloud Connect, which adds capabilities to link SAP’s data cloud with partner platforms like Databricks and Google Cloud. The solution allows data to remain securely inside SAP systems while still instantly accessible in customers’ existing platforms.

AI tools have the potential to unify data silos, resulting in faster, better informed decisions, improved collaboration, and lower costs. Surveyed organizations using AI to break down silos report benefits such as faster decision-making, better visibility, and increased cross-department collaboration.

Some enterprise platforms are embedding AI assistants directly into daily workflows, helping employees in roles like finance, human resources and customer service complete tasks more efficiently. For example, SAP’s Joule assistant surfaces relevant insights in context and automates routine actions across departments.

The Impact of AI

Since the advent of generative AI, some organizations have introduced problems into their business through ineffective AI implementation or misuse of AI tools: Attorneys have filed documents filled with phony case law hallucinated by AI tools; apps built by vibe coders have frequently proven brittle and insecure; and some customer service chatbots have frustrated users with irrelevant or circular responses.

While AI hasn’t delivered a silver bullet, most organizations see positive returns. Those using AI to break down silos cite faster decision-making, improved visibility, increased collaboration, and reduced duplicate work as top benefits. And overall, survey respondents say AI is producing a positive return within their own organizations. In fact, less than 13 percent say they are not seeing value from the AI solutions their organizations have implemented.

Of those that are already using AI to break down data silos, more than 50% say the solutions are resulting in faster decision-making. And more than 40% cite benefits including improved visibility into operations, increased cross-department collaboration, improved customer or employee experience and reduced duplicate work.

Real-world examples show how these benefits translate into measurable business results. For instance, Brazilian food company BRF accelerated demand planning and cut planning time by 33% with embedded AI. Martur Fompak, a Turkish automotive company, streamlined human resources processes and reduced execution time nearly 98%. And Uniper, an international energy company based in Germany, boosted procurement accuracy 95% through automation.

SAP’s new generation of enterprise applications—spanning supply chain, customer engagement, and procurement—illustrates how AI is being built directly into the systems that run day-to-day business operations. By connecting data across these functions, tools like SAP Supply Chain Orchestration and SAP Ariba are designed to help organizations detect risks sooner and coordinate faster responses.

Among readers whose organizations are not yet using AI to break down data silos, the expected benefits are remarkably similar—although they are cited in slightly smaller numbers. Improved decision-making still takes the top spot, cited by just under half of these readers. And at least one-third expect benefits including increased visibility, reduced duplicate work, improved customer and employee experience and increased cross-departmental collaboration.

Asked what opportunities would open up if their organizations could completely break down data silos with AI, readers provided a wide range of responses. One chief revenue officer working in technology, media and communications said this would result in “improved bookings, better retention, cost savings.” An engineering executive from the same industry said: “Product development would happen faster, new markets would open up, teams would have a better sense into what the gaps and challenges are in the market.”

One executive in healthcare and life sciences put the opportunity in simple, compelling terms: “The chance for revenue to explode.”

What’s Holding Organizations Back?

Readers cite a number of obstacles to using AI to break down data silos. However, most of these challenges seem to be related to organizational limitations or the data itself—not to the limitations of AI tools.

Obstacles to using AI for breaking down silos are mostly organizational: poor data quality, privacy concerns, employee resistance and budget constraints. Few cite unclear ROI or lack of appropriate tools as major challenges. Less than 25% of readers cite either unclear ROI of using AI to break down silos or a lack of appropriate tools or technology as a challenge. This seems to suggest readers largely believe that AI tools could provide even more value if organizations first take steps to ensure data quality, implement security guardrails, allocate appropriate budgets and encourage employee adoption.

Many enterprise AI providers, including SAP, are emphasizing responsible AI development — incorporating ethical review processes, compliance with emerging global regulations, and safeguards around data privacy and user oversight that align with global standards like the EU AI Act and UNESCO principles.

Conclusion

• Data silos create damage. Most readers say data silos slow down decision-making in their organizations; silos are also often connected to ineffective collaboration and a lack of business insight. 
• Organizations are already using AI to bring data stores together. More than half of organizations have already put silo-busting AI tools in place or plan to do so soon; and more than 60% say AI tools are providing positive value in their organizations.
• These AI tools are both breaking down silos and producing positive business value. Already, readers cite benefits including faster decision-making, improved visibility, increased collaboration and a better experience for customers and employees. 
• The next step is for leaders to better prepare their organizations to take advantage of cross-functional data access. By tackling challenges like disorganized data stores, data privacy vulnerabilities and employee resistance, organizations can set themselves up to better reap the benefits of AI tools that break down silos.

“As enterprise AI matures, tools that connect systems and data will continue to evolve,” says Khan. “Companies like SAP are expanding AI capabilities to help organizations make faster, more informed decisions, and helping businesses turn data into a real-time advantage.”

Methodology

This report is based on a survey of 106 readers of The Information, conducted in September 2025. Respondents represented a wide range of industries, company sizes, functional areas and titles.

Industry: By far the largest industry represented in the survey was technology, media and communications, which comprised 38% of respondents. Sixteen percent of respondents work for professional services companies, 14% work in financial services or capital markets, and 9% work in healthcare and life sciences. The remaining respondents represented: consumer and retail; industrial manufacturing; hospitality, food, leisure and travel; private equity; and others.

Company Size: Forty-four percent of respondents came from companies with annual revenues under $10 million; 22% had revenues of more than $5 billion; 12% had revenues between $10 million and $100 million; 9% had revenues between $100 million and $500 million; 8% had revenues between $1 billion and $5 billion; and 6% had revenues between $500 million and $1 billion.

Functional Area: Respondents represented multiple functional areas, with the largest groups coming from executive leadership (28%), engineering (13%) and information technology (12%). Other areas, each comprising under 10% of respondents, included marketing and communications, research and development, sales, finance and human resources.

Title: The most represented job title among respondents was CEO and owner (30%), followed by director (19%), employee (16%), manager (10%), senior vice president or vice president (9%), chief information officer or chief technology officer (8%), chief revenue officer (3%) and chief counsel (2%).