数据和人工智能影响力报告的信任必要性

The Trust Imperative What People are Saying AboutThe Data and AI Impact Report “Change is the only constant,and AI is evolving faster thanever, but speed without trustis a risk we can’t afford. Thisreport reveals a striking trend:Industries like banking andinsurance are seeing both highreturns and high volatility – a “Trust is essential to createimpact. This report alignswith our own findings onthe importance of improving Roy IkinkTechnology Lead, Accenture the Netherlands “As an organization deeplycommitted to leveraging dataand AI responsibly, we foundData and AI Impact Reportto be a powerful validationof what drives real businessvalue – trust, governance andscalable data foundations. Thisanalysis goes beyond surface- Preeti ShivpuriPartner, National Leader Trustworthy AI andData Risk, Deloitte “Trying to scale GenAI on weakdata foundations is like buildinga skyscraper on quicksand. ThisThe Data and AI Impact Reportby SAS and IDC unpacksimportant nuances for trust in AIand the impact of AI and offers Ryan BisharaEVP, Revenue & Data Strategy, LAFC and BMOStadium Derek YuehResearch Lead, LinkedIn B2B Institute Contents Executive Summary....................................................................................................1A Message from the Analyst Team.........................................................................2Methodology.................................................................................................................4Introduction...................................................................................................................5The Global State of Data and AI..............................................................................6Core Metrics of this Research..................................................................................7 it delivered the lowest ROI compared to otherAI goals. Organizations with more strategic AIinitiatives significantly expand market share and The success of the data and AI market requiresbusinesses to overcome the trust dilemma. Why is this so important? In just the last two years, generative AI hasalready eclipsed the use of traditional AI. Asthe market rapidly advances to agentic AI, the Agentic AI requires groundwork. The study revealed that agentic AI progress willlikely stall when faced with nonoptimized clouddata environments, poor data governance or Quantum AI is quickly becoming a For the good of society, businesses andemployees: Trust in AI is imperative. Quantum applications are currently beingexplored in logistics, finance, cybersecurity,life sciences, climate modeling and materialsscience. Although this technology is in an In order to achieve trust, the AI industry mustincrease the success rate of implementations,humans must critically review AI results and We set out to tackle these important topics andother questions in a global survey around the In addition to these findings, you will alsolearn about the AI Trust Index and the AIImpact Index – two new measurements we areintroducing to compare the use and impact of AI In these pages, you’ll learn: We tend to overly trust morehumanlike technology. For AI to reach its next growth phase, it mustbe grounded in delivering tangible ROI to In spite of evidence that GenAI can be error-prone, organizations have more faith in thistechnology than in other types of AI, including This requires everyone to overcome the trust Tangible ROI increases with trust. To realize greater value from AI initiatives,the survey found businesses must focus on Businesses must move beyond cost Bryan HarrisExecutive Vice President andChief Technology Officer, SAS Saving money is often a top goal when itcomes to AI initiatives, but the study showed A Message from the Analyst Team Chris Marshall Neil Ward-Dutton Vice President, Data, Analytics, AI,Sustainability, and Industry Research, IDC VP AI, Automation,Data & Analytics Europe, IDC From an academic perspective, the trustdilemma represents one of the mostconsequential barriers to realizing AI’s fullpotential. Our research shows that while 78% oforganizations claim to fully trust AI, only 40%have invested to make systems demonstrablytrustworthy through governance, explainability From a global perspective, but grounded inEurope’s regulatory maturity, it is clear that AI’ssuccess depends less on hype and more on thestrength of its underlying data foundations. Oursurvey shows that weak cloud environments,siloed data and limited governance areconsistently holding back the adoption of all Kathy LangeResearch Director,AI Software, IDC Research Vice President,AI & Automation, IDC What stands out in this research is how rapidlythe center of gravity has shifted from traditionalmachine learning toward generative and agenticAI. Organizations are no longer experimentingon the edges; they are embedding thesetechnologies into workflows that span customerservice, coding and decision support. Yet,