The future will be in the cloud and driven by artificial intelligence (AI), however, a recent Unisys study shows that the reality is more complex. Although 76% of companies plan to increase their investment in cloud infrastructure in the remainder of the year, most face fragmented technology ecosystems that are difficult to scale.
The report, titled From complexity to clarity, modernizing the cloud and technological infrastructure for what’s nextcollects the opinion of more than 1,000 business and technology managers in countries such as Spain, the United States, France, Italy, Germany, Australia and the United Kingdom. Its main finding is that digital ambition is overwhelming the technical capacity of organizations.
Fragmented infrastructure and misdirected resources
The organizations surveyed operate an average of seven different cloud platforms or cloud environments. This dispersion consumes resources that could be used on strategic projects instead of routine maintenance. A clear conclusion emerges from the study: reducing the complexity of the technological ecosystem is an effective way to free up budget for innovation.
While 37% of those running the business point to simplifying infrastructure as their top priority, only 25% of the technology team share that vision.
Recommended tactics to alleviate that complexity include consolidating services, optimizing locations, or combining loads between on-premises and private cloud environments. In fact, the so-called “Innovation Leaders” of the study adopt the hybrid approach as the third most important way to free up budget, combining the public cloud for agility and scalability with on-premises systems to protect sensitive data.
Quality data: cloud and AI
One of the cornerstones of effective generative AI is data quality. The study reveals that 73% of companies plan to allocate a significant portion of their budget to strengthening data management practices. The “Innovation Leaders” place this area in second place in their investment priorities for the next 12 months.
However, only 36% of companies believe their current architecture and tools can support large volumes of data and data-driven decisions. That gap in technical capacity threatens to limit the real return on bets on artificial intelligence.
Security, displaced from the central axis?
Despite technological advances, many organizations continue to react to incidents rather than anticipate them. 85% of respondents admit to having a passive approach to cybersecurity, becoming active only when a threat occurs. The risk is real, with 41% saying each hour of unplanned downtime costs them between $100,000 and $500,000.
However, there are signs of change. According to the report, 62% of organizations have already adopted or are in the process of deploying Zero Trust models, 61% invest in recovery capabilities, and 45% employ managed detection and response systems.
Profile of the “Innovation Leaders”
The study identifies a small group (around 10% of respondents) who firmly believe that innovation should be a strategic pillar. These leaders see security as an enabler rather than a barrier, see the cloud as a source of benefit rather than a cost, and are confident that they will exceed their own innovation expectations within a year.
Although they manage a similar number of platforms as other companies, these leaders prioritize an architecture suitable for each use case instead of adopting generalist solutions. That clarity of vision allows them to move forward with less friction and greater impact.
Challenges beyond technological discourse
The fundamental dilemma underlying the Unisys study is not a lack of desire for innovation, but the disparity between strategic objectives and operational capabilities. Complex infrastructures, fragmented data, and a culture of passive security are slowing down transformations that many companies have already trumpeted.
To close that gap, companies will have to decide whether to focus resources on modernizing the basics, reducing environments, improving data quality and adopting more proactive security models, before launching big bets on autonomous artificial intelligence. Success will depend not only on what is promised, but also on what can be sustained in the medium and long term.
