Artificial Analysis has launched the EnterpriseOps-Gym-AA leaderboard in collaboration with ServiceNow, which evaluates AI agents’ ability to manage real enterprise operations while adhering to business rules. This new benchmark tests agents across eight business domains, such as Customer Service and IT Service Management, using executable SQL checks to verify their final outcomes. Leading the evaluation is Anthropic’s Claude Fable 5 (max), achieving a 51% success rate, closely followed by Google DeepMind’s Gemini 3.5 Flash (high) at 50% and OpenAI’s GPT-5.5 (xhigh) at 47%. The results highlight the models’ varying strengths across different domains, illustrating the challenges of achieving compliance in complex enterprise environments.
ServiceNow: ServiceNow develops a cloud-based platform for enterprise workflow automation, IT service management, and digital business processes, with a focus on integrating AI to connect people, processes, and systems. The company positions its platform as an AI control tower for orchestrating work across organizations. In this news, ServiceNow has partnered to adapt its EnterpriseOps-Gym benchmark for testing AI agents on live enterprise systems through Artificial Analysis.
Artificial Analysis: Artificial Analysis runs an independent platform that benchmarks AI models and API providers on metrics including intelligence, performance, and agentic capabilities. It offers leaderboards and evaluation suites for comparing models on real-world tasks. The organization has launched EnterpriseOps-Gym-AA on its platform in collaboration with ServiceNow to enable outcome-based testing of AI agents in enterprise environments.
AI tools built by Emerald Force
Built and supported by Emerald Force.
You might also like
- When to Ask for JSON for Structured Output
- Fine-Tuning vs. Retrieval: Model Change or Context
- Seeing Isn’t Believing: Verify Video Evidence First
- How AI Can Build Your Case for a Better Raise
- System Prompts: Rules That Govern AI Behavior
- AI Can Draft Books, but Humans Still Make Them Great
- How AI Reviews Code: Bugs, Tests, Style, and Security
- How AI Is Regulated: From Privacy to AI-Specific Laws
- AI Agent Memory: Goals, Preferences, and Past Steps
Value Lock-In: How AI Can Entrench Narrow Values