Akka strives to create more certainty over AI agent’s behavior and decision-making – News

Distributed application development platform Akka, officially known as Lightbend Inc., is launching new tools for artificial intelligence developers that it says will reduce the risks associated with agentic AI systems.

The company believes that one of the biggest challenges of building AI agents is the complex nature of the distributed systems that power them. To fix this, it’s launching a new Akka Agentic Platform that makes it easier to orchestrate teams of AI agents and manage their memory and streaming processes.

Unlike traditional systems that are deterministic, AI agents display emergent behaviors and non-deterministic decision-making, with context-sensitive reasoning. So the same AI agent could make completely different choices on how to deal with the same problem, just because the context of that problem is somewhat different.

This creates a lot of uncertainty around agentic systems, and their unpredictable nature introduces too much risk for some companies to stomach, limiting their potential when it comes to large-scale and mission-critical use cases.

It’s this uncertainty that Akka is trying to tackle with the release of its new Akka Agentic Platform, which features a range of new capabilities that aim to make the behavior of AI agents a bit more predictable.

For instance, there’s an Akka Orchestration feature that helps developer teams to guide, moderate and control multi-agent systems, with support for sequential, parallel, hierarchical and human-in-the-loop workflows. Meanwhile, Akka Agents makes it possible to develop goal-directed AI agents and Model Context Protocol-based tools that can reason, act and analyze, and integrate with third-party agentic systems and applications.

Another key component is Akka Memory, which allows developers to establish a durable, in-memory resource to aid with AI agent context, history retention and personalized behavior. It supports nanosecond writes and is designed to act as both long- and short-term memory, with replication features in case of any system failures.

Finally, the platform includes an Akka Streaming capability for stream processing ambient, adaptive and real-time AI agents capable of continuously processing, aggregating and augmenting live data, video, audio and metrics. These data streams can be ingested from any source and fed into AI agents in real time, keeping them up to date with the world around them.

According to Akka, the combination of these capabilities will bring “complete certainty” to AI agents, so they can achieve the required level of accuracy, safety, availability and recovery that’s needed to deploy them in production.

The company says early adopters have already reported big improvements in their agentic AI projects. Fox Corp.’s video streaming service Tubi said Akka has enabled it to provide customer experiences that are “unlike any other” in the video-on-demand industry, while the Indian food delivery service Swiggy Ltd. claims it was able to reduce the latency of its Swiggy ML platform by more than half.

Akka Chief Executive Tyler Jewell said AI agents require information technology teams to completely rethink their underlying tech stacks. He says they have to adapt from controlling predefined workflows to managing intelligent, adaptive systems operating in open-ended environments.

“Scaling these systems and providing dependable outputs is a tremendous challenge and redefines the meaning of an SLA,” Jewell said. “Akka is unique in that we’re bringing IT the tools to solve this issue at enterprise scale, with enterprise confidence.”

Image: News/Dreamina