Many companies struggle to make their AI projects bigger because it’s hard to connect people, technology, and rules in a safe way. IT teams are in a special spot to help, since they understand both the tech and how to keep things secure. To fix this, IT can bring different teams together, use better tools, and make sure everyone follows the same rules. When IT leads the way, companies can use AI more safely, quickly, and with better results.
What is the biggest challenge enterprises face when scaling AI, and how can IT help overcome it?
The main challenge enterprises face in scaling AI is the orchestration gap – connecting people, processes, and technology securely. IT is uniquely positioned to lead AI orchestration by integrating workflows, ensuring security and governance, and driving cross-department collaboration for scalable, trusted AI innovation.
The orchestration gap most enterprises face
Forrester Consulting’s 2025 report, “Unlocking AI’s Full Value: How IT Orchestrates Secure, Scalable Innovation,” surveyed 417 IT leaders across North America and Europe. The findings show a clear orchestration gap that stalls enterprise AI programs.
- 88% of IT leaders say scaling AI is difficult without orchestration that connects people, processes and technology into one workflow
- 38% cite security and governance concerns as the top blocker keeping pilots from moving into production
- Budget constraints follow at 35%, the inability to prove ROI at 34% and fragmented ownership at 33% (PR Newswire)
Why IT is positioned to close that gap
The same study found that 86% of respondents believe IT is uniquely positioned to orchestrate AI across the enterprise. Their cross-functional visibility and mandate to keep systems secure give them a vantage point no other team enjoys.
“IT sits at the intersection of systems, data, and governance. That vantage point gives IT the enterprise-wide visibility and mandate to balance security with innovation in a way no other function can.”
Key enablers IT leaders highlighted
- Greater visibility across departments – 51%
- Integrated data and workflow platforms – 48%
- Executive sponsorship – 45%
When these enablers are in place, the report notes measurable gains in collaboration, productivity and risk reduction.
Governance and transparency first
Governance and privacy top the priority list yet remain the largest barriers. The study stresses that orchestration embeds security and compliance from day one rather than bolting them on later.
“Orchestration is how IT can overcome these blockers and scale AI with trust, alignment, and impact.”
Practical steps recommended by Forrester:
- Build cross-functional steering committees that include risk, legal and data teams
- Standardize model documentation and version control inside shared platforms
- Use automated policy checks to flag shadow models before they reach production
What leading enterprises are doing now
Forrester’s research aligns with boardroom conversations heard at Technology & Innovation Summits in 2024. Large firms are consolidating data leadership under the CIO, investing in no-code automation and setting up AI Centers of Excellence that report through IT.
A mid-size European bank cited in the study saved 2,400 analyst hours per month by moving fraud-detection models into an orchestrated pipeline. Visibility across departments trimmed investigation time and tightened compliance reporting.
Takeaway metrics at a glance
| Metric | Percentage |
|---|---|
| IT leaders who see orchestration as critical to scale | 88% |
| Respondents confident IT should own AI orchestration | 86% |
| Top blocker – security and governance concerns | 38% |
| Organisations needing better cross-department visibility | 51% |
The data underscores the same message throughout the report: orchestration led by IT transforms AI from isolated experiments into governed, revenue-producing systems.
Why is IT leadership essential for building trust in AI systems?
The latest Forrester study commissioned by Tines finds that 86% of IT leaders believe their teams are uniquely positioned to orchestrate AI across workflows, systems and people. Their enterprise-wide visibility and responsibility for both security and innovation give IT the mandate no other function has to embed governance, compliance and transparency from day one of any AI project.
What specific blockers prevent AI from scaling beyond pilots?
According to the 417 IT leaders surveyed across North America and Europe, the top four obstacles are:
– Security and governance concerns (38%)
– Budget constraints (35%)
– Inability to prove ROI (34%)
– Fragmented ownership across departments (33%)
These barriers keep most organizations stuck at isolated pilots instead of enterprise-wide impact.
How does “orchestration” solve the scaling problem?
Orchestration unifies people, processes, technology and workflows into a connected system, eliminating the siloed ownership and tool sprawl that undermine trust. In practice it:
– Creates end-to-end visibility across AI initiatives
– Builds governance in from the start instead of retrofitting it
– Improves efficiency while lowering security risk
88% of respondents say AI is difficult to scale without such orchestration, underscoring its strategic importance.
What cross-functional model does Forrester recommend to support responsible AI adoption?
The study highlights three prerequisites:
1. Executive sponsorship (45% of respondents call this an enabler)
2. Greater visibility across departments (cited by 51%)
3. Integrated data and workflow platforms (48%)
By combining low-code/no-code automation with clear ownership and shared metrics, IT can shift from reactive support to proactive AI orchestration leadership that aligns every initiative with business objectives.
How are 2025 enterprise roadmaps changing as a result?
Forward-looking companies are bringing data and AI governance directly into IT. Forrester predicts that 30% of large-firm CIOs will fold Chief Data Officer responsibilities into IT this year, ensuring the required data foundations and stakeholder collaboration are in place. Organizations that fail to treat IT as the central orchestrator risk three-out-of-four odds of failure when building advanced agentic AI architectures on their own.
