The vast majority of AI projects are reported to be successful, according to new research that challenges the perception that deployments are likely to fail because the technology is still in its infancy.

In its 2026 AI Automation Benchmark Report, tech firm Jitterbit surveyed 1,500 IT decision makers and found that 78 per cent of respondents said their AI projects had delivered moderate to high value for their organisations.

By contrast, just 2.5 per cent of those surveyed said their AI projects had failed or delivered a negative return on investment.

Where AI projects do encounter problems, a lack of investment has typically been cited as a factor, but Jitterbit reported that this applied to only 15 per cent of IT professionals leading such initiatives.

Security and compliance issues were highlighted as a more significant concern, with 39 per cent of respondents identifying them as primary blockers to AI projects.

With AI deployments facing greater scrutiny from regulators around the world, 47 per cent of those surveyed said AI accountability was the biggest consideration in their choice of AI tools.

Businesses in the software and technology sectors were the most likely to prioritise accountability, with 61 per cent of IT decision makers in these industries describing it as a top priority.

Despite concerns about security and compliance, many organisations are continuing to expand their use of AI.

According to the report, businesses currently have an average of 28 AI agents deployed, a figure Jitterbit expects to rise to 40 over the next 12 months, representing a projected 43 per cent increase.

The findings also suggest that larger organisations are planning to scale AI usage more aggressively. Companies with revenues between £100 million and £499.99 million intend to increase their AI agent deployments from 31 to 49 over the next year, while those with revenues above £500 million are expected to grow from current levels to 72 agents, a 48 per cent increase.

Respondents cited their main objectives for deploying AI tools as bringing products and services to market more quickly (38 per cent), improving customer experience (35 per cent) and reducing organisational technical debt (26 per cent).

Commenting on the findings, Bill Conner, president and chief executive of Jitterbit, said: “The data is clear: the age of the ‘AI pilot’ is over, and the era of the ‘Agentic Enterprise’ has begun.

“Business leaders have moved past budget concerns and are now focused on the strategic imperative of safely and successfully deploying AI at scale.”

---