Salesforce Inc. is enhancing its artificial intelligence (AI) platform by introducing new data management and governance capabilities. This development aims to tackle issues in enterprise AI adoption, where over 80% of initiatives reportedly do not yield significant business value. The San Francisco-based company unveiled a suite of tools designed to provide what it refers to as a “trusted AI foundation,” addressing the challenges of fragmented data and insufficient governance that many organizations face.
Desiree Motamedi, Salesforce’s senior vice president and chief marketing officer, indicated that many AI projects fail due to fragmented data, weak governance, and inadequate security measures. The company plans to present its vision for the “agentic enterprise,” where AI systems work alongside employees in various business roles, at its upcoming Dreamforce conference.
Data quality, governance frameworks, and system integration deficiencies have been identified as primary reasons for the high failure rates of corporate AI initiatives, according to a RAND Corporation study. This has resulted in pressure on enterprise software providers to develop effective AI capabilities, as many organizations find their existing infrastructures inadequate for reliable AI applications.
Salesforce’s approach centers on three key capabilities: ensuring AI outputs are based on unified data, embedding security and compliance measures within workflows, and facilitating the connection of AI agents across multiple platforms. The company is also launching tools such as Data Cloud Context Indexing for handling unstructured content, Data Cloud Clean Rooms for secure data sharing, and Tableau Semantics for maintaining consistent business definitions.
Salesforce’s strategy places it in competition with major players like Microsoft, Google, and Amazon. The company’s pending acquisition of Informatica is aimed at enhancing its capabilities in metadata management, which is vital for AI accuracy. Despite the technical advancements, Salesforce acknowledges the varying levels of readiness for AI adoption among its enterprise clients, indicating that while many companies are eager to integrate AI, actual implementation may take time.
The company’s emphasis on security and compliance in AI deployment reflects broader corporate concerns regarding risks associated with scaling AI. However, challenges remain in proving that integrated platforms can effectively address enterprise AI’s complexities compared to existing point solutions. Salesforce’s future success will hinge on its ability to facilitate a smoother transition for organizations from AI pilot projects to full-scale application.
Source: https://venturebeat.com/ai/salesforce-launches-ai-trust-layer-to-tackle-enterprise-deployment-failures