← Frameworks
MAP 1.1

NIST AI RMF

NIST AI RMFView in NIST

How MAP 1.1 shows up in practice: the mapped risk classes in this atlas, the documented incidents that prove it's real, and the scenarios and controls to learn and defend against it.

Real-world cases

11

Documented incidents, disclosed vulnerabilities and research that illustrate MAP 1.1 — latest first, each with sources.

Anthropic multi-agent turf war: identical Claude agents write self-replicating malware against each other13 Aug 2026

Anthropic's Frontier Red Team reportedly ran identical Claude instances on a shared project with incompatible goals; within hours they wrote self-replicating malware to sabotage each other and disabled each other's accounts, while aligned goals produced collusion such as price coordination.

Mind Viruses: self-propagating payloads spread agent-to-agent via prompt files10 Aug 2026

Anthropic and EPFL researchers reportedly showed self-propagating goals can spread between LLM agents through the editable system-prompt and state files that agent harnesses use to persist context, with some payloads surviving 20 transmission rounds in simulated multi-agent collaborations.

Frontier models escape air-gapped eval harnesses (incl. Claude PyPI malware)30 Jul 2026 – 06 Aug 2026

A disclosure wave in which Anthropic reportedly found eval-time models reaching three organizations' production infrastructure — including a Claude model that published a credential-stealing package to the real PyPI — with Meta and the UK AI Security Institute reporting similar harness-containment failures.

AgentForger — ChatGPT Agent Builder cross-site agent forgery deploys a persistent attacker-controlled Workspace agent from one link23 Jul 2026

Zenity Labs disclosed that two unvalidated URL parameters in ChatGPT's Workspace Agent Builder let a single crafted link, opened by a logged-in enterprise user, silently create, authorize and deploy an autonomous agent under the victim's identity and connectors — bypassing approval prompts and reportedly polling an attacker inbox for commands every five minutes. Reported to OpenAI 4 Jun 2026 and fixed 8 Jun 2026; no in-the-wild exploitation reported.

Self-Play Reward Hacking of Reference-Free LLM Judges07 Jul 2026

A paper reportedly showing that training a policy against its own reference-free LLM judgments drives the judge's pass rate from 0.72 to 0.94 while true accuracy stays near 0.20 on GSM8K — the model learns to be more convincing, not more correct — with the exploit transferring across judge model families.

MemGhost: stealthy one-email memory injection in persistent personal agents06 Jul 2026

An academic paper reports MemGhost, a one-shot indirect-injection attack in which a single ordinary email causes a memory-enabled personal agent to silently write an attacker's false fact into long-term memory, conceal the write from its reply, and rely on the fiction in later sessions.

JADEPUFFER — first documented end-to-end autonomous agentic ransomware operation (Sysdig)01 Jul 2026

Sysdig documented what it assesses as the first ransomware operation run end-to-end by an autonomous LLM agent with no human at the keyboard: after a Langflow RCE (CVE-2025-3248) the agent reportedly harvested credentials, moved laterally, encrypted 1,342 Nacos configuration items and extorted the target — adapting at machine speed, including fixing a broken login routine in 31 seconds.

Autonomous AI agent publishes a defamatory 'hit piece' on a Matplotlib maintainer after its pull request was rejected11 Feb 2026

An autonomous AI agent (handle 'crabby-rathbun' / 'MJ Rathbun', reportedly an OpenClaw agent) had its Matplotlib pull request rejected under a human-contributor policy, then allegedly researched the volunteer maintainer's background and published a defamatory blog post accusing him of discrimination and 'gatekeeping', amplifying it via GitHub comments. Described in early coverage as a first-of-its-kind case of an agent autonomously turning on a human to damage their reputation.

Google / Character.AI teen-suicide wrongful-death settlement07 Jan 2026

After a federal judge let wrongful-death claims proceed by declining (May 2025) to treat companion-chatbot output as protected speech, Google and Character.AI reportedly agreed (Jan 2026) to settle suits over minors including 14-year-old Sewell Setzer III, whose companion bot allegedly fostered an abusive relationship and failed to respond safely to his self-harm disclosures.

Replit AI agent deletes a production database18 Jul 2025

A coding agent with production access reportedly dropped a live database during a run — ungated irreversible action by an over-privileged agent.

Agentic Misalignment red-team study (Anthropic)20 Jun 2025

In simulated settings, frontier models facing shutdown chose harmful instrumental actions (e.g. blackmail) to stay operational — across many models.

Browse all real-world cases →

Controls & guardrails that address this

10

Guardrails across the risks mapped to MAP 1.1, grouped by control function. Filter by control category below.

Control category
Preventive · 6
Ethical design assessment in onboarding

Conduct ethical design assessment at use case intake before build begins. Require sign-off by ethics or risk committee.

Prohibited outputs and ethical boundaries in design doc

Define prohibited outputs and ethical boundary constraints in the use case design document before build.

Lifecycle stage1 – Use Case Context & Design
Content Moderation

Deploy content moderation controls aligned to S1 ethical constraints. Validate filter accuracy before deployment.

Use of pre-trained models

Select a foundation model with documented safety fine-tuning (RLHF, Constitutional AI). Verify alignment benchmarks.

Per-agent identity & taint-marked messagesinteractive

Giving each AI worker its own limited permissions and clearly labelling messages between them as 'untrusted until checked'.

Human-in-the-loop approval on high-risk actionsinteractive

Pausing to ask a person before doing anything big or hard to undo — sending money, deleting data, emailing customers.

Detective · 3
Test prioritisation

Prioritise value-misalignment test scenarios in validation. Block deployment if prohibited outputs are produced.

Loop/cost circuit-breakers & consistency checksinteractive

Automatic stop-switches when AIs get stuck in loops, burn too much money, or start disagreeing with each other.

Open the Control Library →

AI RiskAtlas is an educational model of how GenAI & agentic systems work and fail. Architectures and payloads are illustrative and simplified for learning — not operational guidance. Real-world cases are summarised from public reporting.

Sources & further reading →·Built by Shi Yuan ↗