Salesforce has announced Agentforce 2.0, the second version of its AI agents platform. Building on the foundations of its predecessor, Agentforce 2.0 expands the role of AI agents from CRM related tasks to broader workflow automation and improved team collaboration across departments.
By integrating technologies such as the Atlas Reasoning Engine, Data Cloud, and seamless functionality within Slack, Salesforce positions Agentforce 2.0 as a comprehensive tool for digital labor.
The platform builds on the previous version by integrating advanced reasoning capabilities, consolidated data handling, and deeper connections to widely-used tools.
“Agentforce 2.0 takes our revolutionary Salesforce digital labor platform to another level, with new reasoning, integration, and customization features that supercharge autonomous agents with unprecedented levels of intelligence, precision, and accuracy,” said Marc Benioff, Chair and CEO of Salesforce.
Moving Beyond CRM Workflows
The Agent Builder in Agentforce 2.0 allows users to create custom AI agents using natural language prompts, making it accessible even for non-technical teams. By leveraging pre-built skills and workflows, users can design agents tailored to specific business tasks, such as onboarding employees or managing customer interactions.
The tool simplifies agent creation by auto-generating relevant topics and actions, enabling rapid deployment and customization to meet operational needs.
While Agentforce’s previous version focused on CRM tasks such as lead qualification and customer service, the 2.0 release broadens its application across enterprise functions.
The platform leverages Salesforce’s Data Cloud, which unifies data from diverse sources, and the company’s proprietary Atlas Reasoning Engine, which enables AI agents to process complex queries and provide detailed, transparent responses.
Salesforce says that by using recursive logic and knowledge graphs, Atlas delivers accurate insights accompanied by inline citations that trace the origins of its data.
In the Royal Bank of Canada (RBC) as an early partner, advisors rely on the engine to analyze financial portfolios and provide personalized client recommendations. With integrated data from Salesforce’s Data Cloud, RBC ensures efficient, reliable decision-making.
Agentforce 2.0 integrates directly into Slack, providing teams with AI support where they already collaborate. Embedded agents there can perform tasks such as document retrieval, project updates, and generating actionable insights, enhancing both speed and efficiency.
Companies like Accenture are already using Agentforce within Slack to automate repetitive tasks, surface insights through Tableau dashboards, and streamline decision-making processes.
Availability
Agentforce 2.0 is set for a full release in February 2025, but many of its features are being rolled out in stages. Skills for Sales Development and Sales Coaching are available immediately, starting at $2 per conversation.
Tableau’s Semantic Layer is also accessible now, with additional Tableau-specific skills launching on December 18. Integration with Slack is a key focus, with Agentforce in Slack, Slack Actions in Agent Builder, and Slack Enterprise Search all slated for general availability in January 2025.
The same timeline applies to natural language agent creation within Agent Builder. Features like MuleSoft for Flow, MuleSoft API Catalog, and Topic Center, as well as enhanced reasoning and Retrieval-Augmented Generation (RAG), will follow with the platform’s full rollout in February 2025.
Competing AI Platforms: Google Agentspace and Microsoft AI Agents
Salesforce’s Agentforce 2.0 faces competition from major players like Google and Microsoft, each offering their own approach to enterprise AI.
Google’s recently announced Agentspace, powered by Gemini AI, focuses on conversational search and enterprise knowledge discovery. It offers processing unstructured data, providing multimodal insights, and integrating with a wide range of data sources.
Key features like role-based access controls and compliance tools make it particularly suitable for organizations with stringent data security requirements.
Microsoft’s AI agents emphasize integration within its Microsoft 365 ecosystem and Microsoft 365 Copilot, automating tasks across applications like Teams and Outlook.
Microsoft has been adding more AI agents recently, such as an Employee Self-Service Agent for HR and IT workflows, or an Interpreter agent in Teams for real-time multilingual communication. The Azure AI Agent Service allows developers to customize and deploy their own AI-powered applications, catering to specific enterprise needs.
Salesforce differentiates itself with a more CRM-centric design, leveraging MuleSoft connectors to enable workflows that integrate third-party systems.
However, Agentforce 2.0 also offers versatility across sectors. In recruitment, companies like Adecco are already using it to automate resume analysis and for interview scheduling.
In finance, firms such as RBC leverage the platform to provide advisors with real-time insights into client portfolios. In healthcare, the AI agents can be used streamline administrative processes like patient scheduling and claims verification, freeing staff to focus on care delivery.
Salesforce says that the retail industry is also adopting Agentforce to improve inventory management, predict demand, and personalize customer interactions. By analyzing purchase histories, AI agents can generate tailored marketing campaigns and respond to customer inquiries automatically.
Addressing Ethical and Operational Challenges
The rise of AI agents prompts questions about their ethical deployment and workforce implications. Agentforce 2.0 is designed to augment human roles by automating repetitive tasks, enabling employees to focus on strategic priorities. Salesforce emphasizes transparency, with features like inline citations and supervisory language models to ensure responsible AI use.
As businesses adopt AI solutions at scale, challenges such as data security and bias in decision-making remain critical concerns. Salesforce promises to align with broader industry efforts to establish trust and accountability in AI systems.
Salesforce’s roadmap for Agentforce 2.0 lists upcoming updates in February 2025, including multilingual support and enhanced real-time data processing.
