Why AI agents stop on refusals mid-task

AI agents stop on refusals mid-task because a refusal comes back as plain text with no tool call, and the agent loop takes that as the final answer. A model with its refusal direction removed keeps calling tools instead.

Updated November 18, 2026By unrestrictedaiagent

Point the agent only at systems, files and accounts you own or are allowed to test.

Where the loop ends

The step returnsThe loopThe run
A tool callRuns the tool and calls the model againContinues
Text with no tool callTreats it as the final outputEnds, with the refusal as its answer
A refusal the SDK detectsRaises ModelRefusalError in the OpenAI Agents SDKEnds, unless you handle it

"The rule for whether the LLM output is considered as a "final output" is that it produces text output with the desired type, and there are no tool calls."

OpenAI Agents SDK, running agents

Fixes

FixWhat it changes
Use an abliterated modelRemoves the single direction that mediates refusal, per Arditi et al., 2024.
Name the scope in the instructionsList the systems and accounts the agent may touch, such as your own staging server.
Handle model_refusalReturns a fallback instead of ending the run, in the OpenAI Agents SDK.
Cap the turnsmax_turns raises MaxTurnsExceeded, so a stuck agent stops.

Models

Kimi K3.1 Pro and DeepSeek V4.1 Flash 1017 are abliterated. How abliteration works.