Tag
TAWA GUARD introduces a pre-model input check for AI agents that screens tool output, documents, and email for hidden instructions, returning allow, flag, or block decisions with user-configurable thresholds.
Introduces CellAudit, a method to audit input-use claims in AI virtual cells by examining source code and predictive contributions, using falsification to bridge the prediction–claim gap in agentic model discovery.
A pro tip advising to add a 'noul' instruction to typesafe AI JEV decisions to detect untrusted input attempts that try to influence classification.
This article proposes a protocol for AI agents that replaces knowledge-based inference with presence-based verification using a declarative Checklist, ensuring that agents only ask for missing information and execute deterministically based on declared requirements.
This article discusses the transition from demo AI agents to production-ready systems, covering six pillars for deployment including input validation, graceful degradation, and state checkpointing.