Skip to main content
Mallory
← Back to News
ai-platform-security

Anthropic Warns Single-Turn Safety Tests Miss Multi-Turn AI Misuse

Emerging Technology Risk · 3 sources · Updated 4d ago · First seen Nov 17, 2026

Anthropic Warns Single-Turn Safety Tests Miss Multi-Turn AI Misuse

Anthropic warned that increasingly capable AI models are outgrowing common safety benchmarks, allowing strong test scores to obscure weaknesses relevant to cybersecurity misuse, fraud and other harmful behavior. The company reported aggregate harmless-response rates of 94%–98% in single-turn evaluations on its API, but said multi-turn evaluations produced a wider range of results. Malicious users can gradually escalate requests across a conversation to obtain harmful information that models would refuse when asked directly. Anthropic also identified unrealistic scenarios, contaminated test data, reward hacking and unreliable automated graders as factors that can undermine safety conclusions.

Following unspecified evaluation incidents, Anthropic said it hardened its testing environments, temporarily stopped many internet-enabled evaluations and introduced real-time monitoring for web-search evaluations requiring internet access. It recommended realistic, sandboxed, multi-turn testing with difficulty calibrated to model capabilities, broader coverage and domain-specific scoring rather than reliance on a single aggregate safety score. For enterprise assurance, its guidance emphasizes human validation of automated graders, preregistered analysis, transparent reporting of uncertainty and deployment configurations, and continuous testing throughout model development. Anthropic also called for responsible disclosure of severe flaws in deployed systems.

New surface, new adversariesTrack how attackers are adapting to this technology.
Track this risk
Event timeline

How this story unfolded

Events from the most recent confirmed update back to the earliest known activity.

3 events
  1. Reported Nov 17, 20265d ago

    Anthropic and OpenAI pause model training to strengthen safety measures

    Anthropic and OpenAI paused model training to strengthen safety and alignment guardrails and improve safety-testing procedures, according to the news report.

    Anthropic Warns AI Safety Tests Are Becoming Too Easy
  2. Reported Nov 13, 20269d ago

    Anthropic discloses safety benchmark saturation and multi-turn testing weaknesses

    Anthropic reported aggregate harmless-response rates of 94%–98% on largely saturated single-turn API evaluations, while multi-turn tests exposed weaknesses involving gradual conversational escalation. It recommended realistic, sandboxed multi-turn evaluations, validated scoring methods, and more detailed reporting rather than reliance on aggregate safety scores.

    Building strong safety evaluations \ Anthropic
  3. Reported Nov 13, 20269d ago

    Evaluation incidents prompt Anthropic to harden testing environments

    Following unspecified evaluation incidents, Anthropic hardened its testing environments and temporarily stopped many internet-enabled evaluations. It also introduced real-time monitoring for web-search evaluations that require internet access.

    Building strong safety evaluations \ Anthropic
Linked entities

Related entities

Vulnerabilities, threat actors, malware, products, organizations, and breaches Mallory has linked to this story.

Open in app →

Affected products

1 linked
Claude

Organizations

1 linked
Anthropic
Do you run this?Follow how adversaries are adapting to this technology, and where it touches your stack today.
Track This Risk
Mallory platform

See the full picture, correlated to your attack surface.

This page covers what’s public. Mallory adds the parts that aren’t: which of your assets are affected, which threat actors are using it right now, and what to deploy next.

Exposure mappingMap indicators from this story to your assets and identify affected systems in minutes.
Threat actor evidenceEvery observed campaign, victim, and pivot linked to actors named in this story.
Associated malwareMalware, exploits, and IOCs connected to the activity described here.
Detection signaturesYARA, Sigma, and Snort rules deployed to your SIEM as soon as they’re published.
Scheduled alertsGet matching new stories delivered to your team as they break, not the next morning.
AI threadsAsk questions about this story and take action on the answers.