A new survey by IBM's Institute of Business Value reveals concerns and strategies for the application of artificial intelligence. The survey covers 5,000 executives in 24 countries around the world, and deeply explores the current situation and challenges of enterprises in AI governance, accuracy, bias and risk management. The survey results show that nearly half of CEOs expressed concerns about the accuracy and bias of AI, highlighting the huge pressure companies face in responsibly developing and applying AI technologies. The report also analyzes the differences in AI governance among enterprises with different technological maturity levels and puts forward corresponding suggestions.
Nearly half of CEOs are concerned about the accuracy and bias of artificial intelligence (AI) according to a new survey by IBM's Institute of Business Value. The survey was conducted in collaboration with the Oxford Economic Research Institute and involved 5,000 executives from 24 countries, covering business leaders in North America, Latin America, Europe, the Middle East, Africa and Asia.
In terms of AI governance, the survey shows that 21% of executives believe that their organization's AI governance maturity is in a systematic or innovative stage, which shows that companies still have a lot of room for improvement in this area. AI governance refers to principles, policies and responsible development practices that are consistent with ethics and human values.
To address the concerns posed by AI accuracy and bias, 60% of C-suite executives said they had set up clear generative AI leaders within the organization. Meanwhile, 78% of executives said they maintain detailed documentation to ensure the interpretability of AI. In addition, 74% of businesses conduct ethical impact assessments and 70% conduct user testing to assess and mitigate potential risks. The survey also found that 80% of C-suite executives said that companies have dedicated risk management functions that focus on the application of AI or generative AI.
IBM Consulting Global Leader for Trust AI Phaedra Boinodiris pointed out that establishing a strong governance framework that promotes accountability, transparency and interpretability is the main focus of enterprises at present. She suggests that when companies establish a responsible AI governance foundation, they can consider a series of actions, including improving the AI literacy of all employees so that they can not only have the technical skills to effectively use AI, but also develop critical thinking skills. In addition, companies should ensure that their measurement systems are aligned with core values, including the values of their stakeholders, while introducing diverse and multidisciplinary teams into the governance system for AI model development and procurement.
The research also shows that companies with high technological maturity often pay more attention to AI governance, while companies with low technological maturity face the complexity of governance choices. IBM stresses that a flexible AI governance framework can help companies adapt to market changes, reduce risks, and facilitate greater adoption to realize the potential of AI.
This research report is one of IBM's Business Value Institute's series on generative AI, aiming to reveal the opportunities and challenges this technology brings to enterprises around the world. Other related reports show that 77% of business leaders believe that generative AI is ready to enter the market and that rapid adoption of this technology is crucial to stay competitive.
Key points:
Nearly half of CEOs expressed concerns about the accuracy and bias of AI, and companies still have room for improvement in governance.
60% of enterprises set up generative AI leaders, and 78% maintain detailed documentation to ensure interpretability.
High-tech mature enterprises attach more importance to AI governance, and a flexible governance framework can help reduce risks and adapt to market changes.
In short, this survey by IBM provides an invaluable reference for enterprises in AI governance and risk management, emphasizes the importance of building a responsible AI ecosystem, and points out the direction for future development. Enterprises need to actively respond to the challenges and opportunities brought by AI to ensure the healthy development and sustainable application of AI technology.