持久人工智能准备：如何构建和维持数据治理项目.pdf

《持久人工智能准备：如何构建和维持数据治理项目.pdf》由会员分享，可在线阅读，更多相关《持久人工智能准备：如何构建和维持数据治理项目.pdf（22页珍藏版）》请在三个皮匠报告上搜索。

1、Durable AI Readiness:How to Build and Sustain a Data Governance ProgramNINA CARTER|PRESIDENT,THE INFORMATION PROFESSIONALSAgenda01 Reframing the Conversation02 The Evidence AI Readiness Research03 What a Data Governance Program Looks Like04 A Real-World Build05 Before You Begin06 Making It Operation

2、alNina Carter,President The Information Professionals+30 yrs.Experience Information Governance,Data Governance,AI ReadinessAIIM Board of Directors Member,Chair Membership CommitteeFirst EDRM Solution Deployment Foremost-(1993)400+clients Subscription,Legislative ServicesPublic and Private Sector Cus

3、tomers throughout North AmericaTHREE UNCOMFORTABLE TRUTHSLets Start Here80%+RAND Corp,2024of AI projects fail to reach production twice the failure rate of traditional IT projects73%Industry benchmarks,2024of leading organizations have formal data governance programs with dedicated staff and funded

4、roles$0The real question is the cost of waitingis the cost of starting today vs.audit exposure,legal risk,and reputational harm from ungoverned AIThe question is not whether to govern your data.The question is how long you can afford not to.REFRAMING THE CONVERSATIONData Governance Is Not What Most

5、People ThinkCOMMON MYTHIts an IT projectREALITY:Most data quality problems are because of a lack of defined processes.COMMON MYTH“Having a data governance program is a new trendREALITY:Formal data governance frameworks predate the internet era 1980s.COMMON MYTH“We can clean our data after we launch

6、AIREALITY:Poor data quality is a liability and security issue an ungoverned model trained on bad data can make consequential decisions at scale.Two Organizations.Same AI Ambition.Opposite Outcomes.WITHOUT DATA GOVERNANCE WITH DATA GOVERNANCEDATA QUALITYInconsistent,siloed,untrusted.Nobody agrees on