Salesforce Agentforce 3 aims to tackle what many companies are struggling with. It’s about seeing what the AI agent actually does.
Since debuting in October 2024, AgentForce has achieved several victories in various sectors. The engine was able to reduce customer case handling time by 15%, but 1-800 Accantant passed 70% of the management chat queries to AI during the insanity of tax season.
But what’s interesting about this upgrade is that not only numbers, but no one wants to talk about how the Salesforce deals with the elephants in the room.
Keep agent tabs
The AgentForce 3 centerpiece is what Salesforce calls a command center (essentially mission control for AI employees). The manager can peer under the hood to find patterns in the agent’s performance, track real-time (latency, escalation rate, errors), and identify which bits and swift kicks are needed.
For those who deploy AI tools and wonder, “What’s it?”, this level of visibility can change the game. This system uses the OpenTelemetry standard to capture all agent activity. This means it works well with tools like Datadog and Splunk that your IT team probably already have on the screen.
AI adoption is absolutely rising. Future data from the Slack Workflow Index shows that AI Agent usage has increased by 233% in just six months. Meanwhile, around 8,000 organizations have signed up to deploy AgentForce.
Ryan Teeples, CTO of 1-800Accountant, said:
“We have established a powerful deployment foundation, and each week focuses on launching new agent experiences and AI automation through the latest features of AgentForce.
Salesforce AgentForce 3 not only provides data, but it actually suggests improvements. AI effectively monitors itself, identify conversation patterns and recommends fine-tuning. It’s a bit meta, but it’s very useful for teams who don’t have time to manually review thousands of bot interactions.
Have you solved the connectivity challenge?
Another headache-filled sales force is connectivity. AI agents are just as convenient as accessible systems, but getting a firm connection to business tools has been a pain for most organizations.
AgentForce 3 brings native support for the Model Context Protocol (MCP). This is rather appropriately described as “AI USB-C.” This essentially means that AI agents can connect to any MCP-compliant server without custom coding, while respecting their security policies.
This is where Mulesoft (acquired by Salesforce a few years ago) works to convert APIs and integrations into agent-enabled assets. Heroku will then handle the deployment and maintenance of your custom MCP server.
Mollie Bodensteiner, the SVP of Engine, commented: “Salesforce’s open ecosystem approach helps us fully and confidently scale the use of AI agents, particularly through native support for open standards such as MCP.
“You can securely connect agents to enterprise systems that rely on agents without custom code or governance breach. That level of interoperability gives them the flexibility to accelerate adoption and keep them fully in control of how agents behave within their environment.”
Growth of the Salesforce AgentForce ecosystem
Perhaps the most interesting aspect of this announcement is the ecosystem they are cultivating, not something Salesforce built themselves. Over 30 partners have created MCP servers that integrate with AgentForce, including players such as AWS, Google Cloud, Box, PayPal, and Stripe.
These integrations go far beyond simple data access. For example, AWS integration allows agents to analyze documents, extract information from images, transcrib audio recordings, and identify critical moments in the video. Google Cloud Connections is tied to maps, databases and AI models such as Veo and Imagen.
Healthcare appears to be a particularly promising sector.
Tyler Bauer, VP of Uchicago Medicine’s Systems Outpatient Business, explained:
“We need to support that goal by automating daily interactions in patient access centers, including general questions and requests.
The real question of course is whether all of this actually helps to manage the growth force of AI agents deployed by the company. Visualizing AI performance has been a blind spot for many organizations. Many times we know a large percentage of queries our AI is handling, but we struggle to identify specific drawbacks and opportunities for improvement.
Adam Evans, EVP & GM at Salesforce AI, said: “AgentForce 3 redefines how humans and AI agents work together.
It remains to be seen whether it will withstand that lofty promise, but addressing the gap between vision and control is a step in the right direction for businesses struggling to properly manage their AI initiatives.
See: Huawei Harmonyos 6 AI Agent offers an alternative to Android and iOS
Want to learn more about AI and big data from industry leaders? Check out the AI & Big Data Expo in Amsterdam, California and London. The comprehensive event will be held in collaboration with other major events, including the Intelligent Automation Conference, Blockx, Digital Transformation Week, and Cyber Security & Cloud Expo.
Check out other upcoming Enterprise Technology events and webinars with TechForge here.
