Amazon now lets AI assistants start the business workflows you already built on AWS, with approvals and rules kept intact at every step.
Amazon Web Services now lets AI assistants start the business workflows you already built on its platform. The AI decides when to run a workflow and what information to hand it. Then a service called AWS Step Functions carries out each step exactly as you defined it.
Step Functions is a visual tool for lining up the steps of a process, and it can coordinate over 220 AWS services. Amazon Bedrock AgentCore Gateway is the connector that links AI agents to tools, using a standard called the Model Context Protocol. Agents can use both Standard and Express workflows, the two existing types.
AWS gives a refund example. A customer support agent starts the refund workflow. Step Functions then runs the eligibility check, the manager approval and the credit, so every refund follows company policy. Agents can also look up past runs. An operations agent could tell an on-call engineer which order runs failed overnight and at which step.
You pay standard rates for Step Functions and AgentCore Gateway. AWS mentions no separate fee, and it gives no price figures here. It says the feature works in all Regions where AgentCore Gateway is available, but it does not list them. A tutorial is in the Step Functions Developer Guide.
If your business processes already run on AWS, an AI agent can now trigger them while your approvals and rules still apply at each step.
Otto, a new customer-tracking app for field sales teams, lets reps speak their notes instead of typing them. Price and launch details are not yet known.
Read more: Otto launches a sales app you talk to instead of typingWorld models, AI that learns physics from video, are being built into home robots and driverless cars, and your own videos may help train them.
Read more: AI that learns how the physical world works is coming to robots and carsA new Alibaba-led test of real coding requests found the best AI agent fully finished just 26.4 percent of tasks, so human review still matters.
Read more: AI coding helpers fully finish only about 1 in 4 real tasks
