Artificial intelligence (A.I.) has commanded attention across industries, but many leaders are grappling with how to turn its promise into actual progress. While executives confront skepticism due to reports of slow revenue gains, organizations are re-examining their digital infrastructures, emphasizing data quality and targeted objectives. As interest in A.I. spending grows, businesses are assessing not only technology, but also foundational practices, governance, and regulatory issues. Decision-makers face choices about where to allocate resources to see measurable benefits from new A.I. initiatives. These efforts reveal that meaningful adoption depends as much on readiness and strategy as the power of A.I. itself.
Earlier discussions in public forums frequently highlighted the overall failure rate of enterprise A.I. and the difficulty firms have experienced in integrating these technologies effectively. Emphasis often rested on the technological aspects, rather than the organizational and data challenges. In current developments, the dialogue has shifted to recognize the complexity of existing legacy systems and the critical role that clean, consolidated data plays in achieving real business outcomes. Companies are increasingly seeking not only technical solutions, but also guidance on compliance and ethics, reflecting lessons learned from previous rollouts.
What hinders A.I. from realizing business value?
Recent industry surveys counter the narrative that A.I. underperforms, noting that failures often stem from outdated infrastructure and poor-quality data rather than from the technology itself. Adam Gabrault, CEO of Solvd, points out that significant benefits are observed when A.I. is deployed within specific departments, from predictive analytics to HR. He stresses the importance of starting with a “think big, take small wins” approach, where the focus is tied to explicit business objectives rather than trying to apply A.I. to every problem at once.
“There’s a huge amount of pressure from all sectors, and all industries, to figure out how A.I. could be a change agent,”
Gabrault stated. Deployments on aged systems can stall progress, highlighting the need to invest in updating underlying data infrastructure.
How do data challenges undermine A.I. efforts?
Messy data and siloed information represent substantial obstacles to wider A.I. adoption in large enterprises. Bakul Banthia, co-founder of Tessell, emphasizes that
“most A.I. initiatives actually die, not from bad algorithms, but from the unglamorous reality of messy data and systems that weren’t designed to share information.”
Improving data consistency requires connecting disparate systems and leveraging automated tools, but also requires human oversight to maintain data quality. Companies that prioritize building an A.I.-ready data stack have observed notable improvements in their deployment outcomes.
Should organizations prepare for increased governance and regulation?
A.I. governance lacks standardized frameworks, leaving companies to navigate new responsibilities around privacy, transparency, and compliance. Attention from regulators is increasing, particularly concerning practices like A.I.-washing, which can result in legal and reputational risks. Some U.S. states and European jurisdictions are introducing legislation, and the EU AI Act is adding further complexity for companies with cross-border operations. Those with robust compliance structures are better positioned to adapt, particularly in sectors such as finance and healthcare where regulatory scrutiny is most intense.
Organizations evaluating A.I. must focus on upgrading digital foundations and confirming the integrity of their data as a precondition for adoption. Case studies consistently show that measurable gains arise when organizations define target outcomes, modernize their infrastructure, and prepare for regulation. Ethical considerations and transparent governance frameworks will also play a growing role, especially as enforcement intensifies. Beyond the technology, steady discipline in planning, implementation, and compliance can help companies overcome barriers and move beyond initial hype toward tangible value. Formulating concrete project goals, investing in data quality, and involving compliance experts early will be key steps for enterprises planning their next moves with A.I. For managers, these lessons point to the necessity of cross-functional teams and continued investment in training, so that A.I. becomes an asset rather than a liability.
