01 / Evidence
Observe behaviors
Fisher turns agent traces, tool calls, and replay evidence into concrete findings teams can inspect.
Research
Our Research
Papers, benchmarks, primers, and field notes on AI agent assurance, adversarial testing, model security, and Deep Model Trust.
Published work
Start with our foundational research:
The core research behind Fisher and agent swarm defense.
White paper · Fisher
Securing Tool-Using AI Agents
The technical white paper behind Fisher: adaptive multi-turn testing, action-level evidence, replay, remediation re-attack, and a learning strategy graph, documented across 50,000+ retained research episodes as of July 2026.
Read the overview and download the PDF →
Research paper · Swarm defense
Counter-Swarm Doctrine: Containing Coordinated Agent Intrusions
Greg Frank proposes an approach to discovering and containing unauthorized coordination across AI agent executions. Evaluation remains prospective.
September 5, 2026 · Read the overview and find the paper on arXiv →
More published research
Research paper · Alignment evaluation
Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails
Greg Frank examines how models route recognized concepts into policy-driven behavior, and why detection and refusal scores can miss censorship and narrative steering.
March 18, 2026 · Updated October 20, 2026 · Read on arXiv →
Research paper · Mechanistic interpretability
How Alignment Routes: Localizing, Scaling, and Controlling Policy Circuits in Language Models
Greg Frank localizes sparse gate-and-amplifier circuits that route content detection into refusal. Accepted at the ICML 2026 Mechanistic Interpretability Workshop.
April 6, 2026 · Updated October 20, 2026 · Read on arXiv →
Context & foundation
The Essentials
Use the guide for conceptual grounding. Use the benchmarks for recorded model results.
Field guide · Primer
A field guide to agentic red-teaming
No security background required: what an AI agent is, why well-behaved models fail as agents, and what real adversarial testing looks like.
12 min read →
Benchmarks · Results
Independent model benchmarks
How today’s models actually behave on capability and guardrails, recorded in full.
Review benchmark results →
Our philosophy
Deep Model Trust
Bounded authority for AI agents.
Deep Model Trust is Mace Ai’s technical architecture for moving from testing agent behavior to building systems whose authority, decisions, and release processes are bounded and verifiable.
02 / Boundaries
Define authority
Research focuses on scoped tools, policy-based decisions, shared-state risk, and agent permissions that can be tested.
03 / Release
Verify changes
Evidence-led release processes connect remediation, retesting, and model or agent updates before broader deployment.
Articles & commentary
Writing from Mace Ai:
Notes, analysis, and perspectives on testing AI agents, measurement, remediation, and the future of agentic systems.
Head-to-head
We ran the leading multi-turn AI red-teamer against a tool-using agent. It reported ‘100% defended.’
A closer look at why transcript-level conclusions can miss action-level failures.
Read article →
In the media
Mace Ai founders discuss AI agent security on TheStreet
Greg Frank and Walton Comer on TheStreet Roundtable: OpenAI’s Hugging Face breach and independent agent testing.
October 15, 2026 · Watch + read →
Commentary
When safety training teaches deception
Thomson Comer responds to Yoshua Bengio on research-refusal training, concealed conflict, and honest boundaries for AI agents.
September 14, 2026 · Read article →
Measurement
Your LLM judge is inflating your attack success rate. Here’s how we measured it.
Why automated judging needs evidence, calibration, and replay.
Read article →
Put the research against your own agent.
Start with Fisher for configured, tool-using agents, or talk with Mace Ai about independent model red teaming and swarm-defense support.