MuleSoft integration platform from Salesforce Inc. Today announced support for a new event-driven application programming interface tool that will allow businesses to create artificial intelligence agents with real-time data.

Now generally available, MuleSoft support is intended Asynchronous APIa standard open source initiative designed to provide connections to deliver real-time data through event-driven architectures. Using this standard, AI systems can react to changes in internal and external systems, such as data changes, business rules, predefined automations or other signals.

MuleSoft said real-time events are common for transmitting data to customers and users. However, the company cited a recent report revealing that 43% of IT leaders have failed to integrate traditional systems with their EDAs.

Salesforce recently released Agent forcea suite of tools for creating and customizing AI agents. Unlike traditional chatbots, AI agents can act by themselveswithout human intervention, and can be triggered by events. The integration of Simpler methods of consuming events provide enterprise developers with a key tool for building better agents.

Using a tool like AsyncAPI allows AI agents to work asynchronously, meaning they don’t need to request an update and wait. Instead, an agent can log into an event-driven system during its process, request to be updated if anything changes, and continue its work.

Andrew Comstock, vice president of product at Salesforce, told SiliconANGLE in an interview that this makes this very powerful for AI agent use cases.

“If you think about the path development has worked in the past, most APIs are measured in microseconds and milliseconds,” Comstock said. “Well, an agent can perform a very complex and much longer task.”

Comstock explained that by using the new standard, companies would be able to Reliably rely on their event-driven architectures to access real-time data streams. Agents would then be able to act based on changing information and adjust their responses in real time as soon as an event occurs. For example, if some important data collected earlier in the process were updated, it would affect the agent’s reasoning – in the same way that people adjust their thinking when new information arrives.

The other automation use case allows AI agents to be activated by real-time events with standardized information provided by the API and data source so they can take immediate action . “This is where the rubber meets the road in agent desire,” Comstock said.

Many industries would benefit from this standardization and the incorporation of real-time adjustments. For example, a customer support agent might work on a call to resolve an inventory issue when new stock arrives while the agent processes the information in the background. If the warehouse database is updated while the agent is working, this resolves the issue, he can quickly change course. Another agent may be required to take action when certain inventory thresholds fall below certain levels and notify team members or call in new orders.

Real-time event data allows AI agents to take automated actions to reduce risks and prevent service interruptions in IT systems. For example, an agent monitoring compute with historical data on usage spikes could receive events related to sudden increases in resource usage and operate backup servers without human intervention.

Comstock explained that many businesses that didn’t have access to or the skills for this technology before have started to expand now that the tools are more available.

“By enabling products like MuleSoft, we allow them to access technology spaces that they might not have been able to access otherwise,” Comstock said. “I see this as a potentially important unlock that maybe they didn’t feel comfortable in or didn’t feel like they had the skills. Using a tool like MuleSoft to complement their skillset is an enabling capability to enable these types of agentic AI use cases and integrations.

Image: SiliconANGLE/Microsoft Designer