The AI running inside large companies has stopped waiting for permission. Across five categories of AI, from chatbots to process automation, more than half of 3,600 business decision-makers say the systems their teams use can act within defined rules without routine review, or finish whole processes with no human approval. 87% of those same respondents […]
The AI running inside large companies has stopped waiting for permission. Across five categories of AI, from chatbots to process automation, more than half of 3,600 business decision-makers say the systems their teams use can act within defined rules without routine review, or finish whole processes with no human approval. 87% of those same respondents say their organization needs a better way to approve, monitor and control how AI gets used.
Vanson Bourne ran the survey across North America, EMEA, and Asia Pacific and Japan. Every respondent came from a company with more than $500 million in annual revenue, spread across seven sectors including financial services, healthcare, manufacturing and retail.
“There is a dangerous gap opening up between AI confidence and AI reality. Confidence is not the problem. Ambition is not the problem. The problem is that most organizations are using AI to improve a business model that AI should be replacing,” said Chris Howarth, CEO at Avanade.
More than half say their AI can act unsupervised
Conversational AI and virtual agents top the autonomy list at 59%. Standalone generative tools, the ones that write text or code, sit lowest at 54%. Predictive analytics and machine learning, robotic process automation, and natural language processing each land at 56%. These are managers reporting on their own teams, so the figures capture what they believe is running.
87% say AI initiatives need faster authorization to stay competitive, and 91% of that same group say the approach to approving, monitoring and controlling AI needs to be more effective. The people pushing to shorten the sign-off are the people who do not trust the sign-off.
Companies that moved fast are already losing track
Among those that have deployed or scaled AI at speed, 46% report security, privacy, regulatory or compliance problems and 25% report loss of visibility, control or agent permissions. The second figure matters most, because an agent whose permissions a team cannot account for is unmanaged access to whatever that agent touches.
45% believe their people use AI output without checking its accuracy or suitability, 32% believe AI is being used in ways their compliance team would not approve of, and 31% report unapproved tools in use. Another 37% say AI has confused who owns a decision.
A mid-level manager at an enterprise healthcare company in Canada, asked how they use AI in ways their employer may not know about: “I realize that AI doesn’t offer the same security as our main systems. But what is the way forward, blocking access doesn’t work. I do upload our business and company information to my AI app. I understand the security implications.”
The governance gap
On AI security, 30% say strengthening is needed and another 40% say moderate. On governance and risk, 94% say further development is needed. 36% put security, governance or regulatory concerns among the biggest barriers stopping their company from becoming an AI leader. Security, privacy, governance and regulatory compliance is also the single most common area of heavy AI investment at 47%, so the spending has started even where the controls have not landed.
Among companies between $500 million and $5 billion in revenue, 90% say a better approach to approval, monitoring and control is needed, against 70% of companies above $5 billion.
Strong governance, access controls and responsible use rank among the highest priorities for 45% of respondents, well ahead of the 26% who name deploying AI as quickly as possible.
