diff --git a/src/pipecat/adapters/services/bedrock_adapter.py b/src/pipecat/adapters/services/bedrock_adapter.py
index 3e99edf6e..2e5c2c62a 100644
--- a/src/pipecat/adapters/services/bedrock_adapter.py
+++ b/src/pipecat/adapters/services/bedrock_adapter.py
@@ -29,7 +29,7 @@ from pipecat.processors.aggregators.llm_context import (
 class AWSBedrockLLMInvocationParams(TypedDict):
     """Context-based parameters for invoking AWS Bedrock's LLM API."""
 
-    system: Optional[str]
+    system: Optional[List[dict[str, Any]]]  # [{"text": "system message"}]
     messages: List[dict[str, Any]]
     tools: List[dict[str, Any]]
     tool_choice: LLMContextToolChoice
diff --git a/src/pipecat/services/aws/llm.py b/src/pipecat/services/aws/llm.py
index 781ae6322..b201f43aa 100644
--- a/src/pipecat/services/aws/llm.py
+++ b/src/pipecat/services/aws/llm.py
@@ -815,14 +815,10 @@ class AWSBedrockLLMService(LLMService):
             messages = []
             system = []
             if isinstance(context, LLMContext):
-                # Future code will be something like this:
-                # adapter = self.get_llm_adapter()
-                # params: AWSBedrockLLMInvocationParams = adapter.get_llm_invocation_params(context)
-                # messages = params["messages"]
-                # system = params["system_instruction"] # [{"text": "system message"}]
-                raise NotImplementedError(
-                    "Universal LLMContext is not yet supported for AWS Bedrock."
-                )
+                adapter: AWSBedrockLLMAdapter = self.get_llm_adapter()
+                params: AWSBedrockLLMInvocationParams = adapter.get_llm_invocation_params(context)
+                messages = params["messages"]
+                system = params["system"]  # [{"text": "system message"}]
             else:
                 context = AWSBedrockLLMContext.upgrade_to_bedrock(context)
                 messages = context.messages