Deloitte: 74% of companies have already achieved or exceeded their overall AI initiatives (but challenges remain)

Companies of all sizes around the world are trying to understand generative AI and identify areas where it can add value. The good news: the majority of organizations actually succeed in achieving this.

According to a new report today from Deloittethe majority of organizations are already meeting or exceeding their own expectations for return on investment (ROI) from AGI. The Q4 State of Generative AI report, based on a survey of 2,773 leaders in 14 countries, highlights the progress and challenges organizations face in their AI journeys.

The report shows significant progress First version Released a year ago, business leaders expressed multiple concerns. There is also positive progress over Third quarter reportwhich showed that the majority of organizations avoided some AGI use cases due to data issues.

Despite taking longer than expected to assess value, nearly three-quarters (74%) of survey respondents reported that their most advanced general AI initiatives meet or exceed ROI expectations. Cybersecurity and IT jobs lead the way in terms of return on investment and successful expansion.

Key findings include:

• Organizations need at least 12 months to solve key adoption challenges
• Information technology, cybersecurity, operations, marketing and customer service show the strongest adoption and results
• Regulatory compliance has emerged as the biggest barrier to the deployment of AGI
• 78% of respondents expect to increase their overall spending on AI in the next fiscal year

Jim Rowan, head of AI at Deloitte, told VentureBeat that the biggest gains companies realize from using AI are efficiency and cost savings.

“We take time away from daily tasks and activities and make people more efficient,” Rowan said.

The challenge is that AGI moves at enterprise speed

Enterprise technology, by definition, is about stability and flexibility. They’re supposed to be things that companies work on. For many types of technology, enterprise adoption can take several years as organizations first need to validate use cases and ROI potential.

While the rapid progress in General AI capabilities It has captured the public’s imagination, and companies often move at a much slower pace when it comes to adoption. This disconnect between the breakneck speed of AI innovation and the more intentional nature of enterprise technology deployment represents a major challenge.

“Companies move at enterprise speed,” Rowan said. “This emerges in two different areas within the report, expanding the range of regulatory questions, risks and challenges faced by organizations across the board.”

This speed disparity is further complicated by the fact that many organizations are still grappling with core technology challenges, such as data management and platform modernization. These fundamental issues must be addressed before companies can fully leverage the potential of generative AI, Ruane noted.

Rather than rushing to deploy the latest AI tools, Rawan emphasized the importance of a more measured, strategic approach that focuses on building the necessary infrastructure and cultural readiness. By taking the time to properly integrate AGI into existing processes and workflows, organizations can ensure that the technology delivers tangible, long-term value, rather than merely acting as a passing novelty. This patient and intentional approach, although likely slower in the short term, may ultimately prove more effective in driving lasting transformation.

AI offers organizations the greatest return on investment today

One of the key areas where organizations see tangible value from AI is in the software development lifecycle.

According to the report, AI helps drive efficiency gains across the entire process – from requirements gathering to testing and deployment.

“We see this a lot in the software development lifecycle,” Rowan said. “That’s why IT has been a big supporter of this.”

Besides software development, companies are leveraging AI to enhance customer service and call center operations. By automating certain tasks and interactions, companies are able to improve efficiency and responsiveness. “The other big use case is around call centers, customer service, and kind of sharing from those two,” Rowan said. “So those tend to be the biggest areas where we see the most efficiency.”

How organizations can measure the impact of AGI

As companies seek to measure the impact of their investments in artificial intelligence, Rawan stressed the importance of looking at quantitative and qualitative measures.

While cost savings and efficiency gains are important, companies must also track the number of new ideas and use cases generated, as well as the impact on employee skills and culture.

In quantitative categories, Rowan cited some key metrics:

• Measure efficiency through cost savings
• Increase revenue generation
• Increased efficiency per full-time equivalent (FTE) employee in certain activities.

On the qualitative side, Rawan pointed to metrics related to employee development, continuous learning, and comprehensive transformation of business operations.

“How are your employees upskilled? How do you use this moment to change the culture around learning and development?” he said.

Leverage the promise of agentic AI

Perhaps the biggest area of ​​innovation that companies should consider in 2025 is agentic AI.

The report notes that 52% of organizations are pursuing AI agents, with 45% specifically exploring multi-agent systems. Ruan expressed optimism about the potential of agentic AI, but noted that it will take time for companies to fully adopt and integrate this technology. He explained that companies are likely to start with simpler, more focused proxy applications before expanding their use.

Agent AI has the potential to radically transform an organization’s operations and deliver a significant return on investment, but only if approached strategically, Rowan said. With initial AI generation deployments, organizations often focus on proof-of-concept (PoC) deployments. A different approach will be needed for agentic AI. Instead of looking at individual use cases, organizations would be well served by looking at the broader process chain. He explained that the real value of agentic AI will come from rethinking entire business processes to be driven by AI, rather than just implementing individual use cases.

“In order to be effective, you actually have to think about how you rebuild operations with the idea that all of this is going to be AI-driven, not human-driven,” he said.

Overcoming the challenges of adoption

Despite the clear benefits, companies still face significant hurdles in scaling up their AI deployment.

One major barrier, according to Deloitte, is limited access to and use of AI tools within the workforce. According to the report, less than 40% of the workforce in most organizations has access to general AI tools.

This lack of widespread adoption indicates the need for a cultural shift, where employees are not only given the tools, but understand the value and importance of integrating AI into their daily workflow.

“If you’re not using AI once a day in your daily life, whether it’s a corporate tool you’re given or a consumer-based tool, I think you’re missing out,” Rowan said.