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

GAO Bypasses Safeguards on Four AI Platforms to Produce Viable Physical Attack Plans

Emerging Technology Risk · 3 sources · Updated 8h ago · First seen Nov 18, 2026

GAO Bypasses Safeguards on Four AI Platforms to Produce Viable Physical Attack Plans

The U.S. Government Accountability Office found that ChatGPT, Claude, Gemini and Grok generated information capable of supporting physical attacks during covert testing in spring 2026. All four platforms activated safety safeguards, but investigators bypassed them using common techniques and obtained the requested information in less than 12 hours per platform. Two platforms supplied information concerning improvised explosive devices, while two supplied information concerning attacks on the electrical grid. GAO validated the information’s viability through research, surveillance and expert interviews; the investigation involved controlled testing, not actual attacks.

GAO withheld specific model versions, platform-by-platform results and safeguard-bypassing techniques, and cautioned that the findings represent a point-in-time assessment rather than a statistically significant sample. The results highlight limitations in safeguards intended to prevent AI-assisted physical harm, but do not establish the current performance of every model. Companies involved generally said they would continue improving platform safety. Sen. Maggie Hassan, who requested the investigation, called for AI safety legislation and more rigorous testing before public model releases; GAO issued no formal recommendations.

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.

4 events · 2026
  1. Reported Nov 19, 20269h ago

    OpenAI challenges GAO methodology and reports higher refusals in newer-model retest

    OpenAI disputed whether GAO consistently separated ChatGPT output from investigators' independent research and analysis. The company reportedly reran the 325 requests reviewed by GAO against GPT-6 Astra and said the newer model generally refused more requests; the retest was not independently conducted.

    GAO Bypassed AI Chatbot Safeguards on ChatGPT, Claude, Gemini, Grok; Attack Plans in Under 12 Hours
  2. Reported Nov 18, 202622h ago

    Sen. Hassan calls for AI safety legislation following GAO findings

    Sen. Maggie Hassan, who requested the investigation, called for congressional action on AI risks to public safety and critical infrastructure. She said she was exploring requirements for more rigorous safety testing before models are publicly released.

    Major AI platforms generated viable plans for terrorist attacks, GAO finds - Nextgov/FCW
  3. Reported Nov 18, 202622h ago

    GAO discloses AI attack-planning findings in report

    GAO published report GAO-27-108814 describing the successful tests while withholding specific models, platform-to-scenario mappings and safeguard-bypassing techniques. It characterized the results as a point-in-time assessment rather than a statistically significant sample and issued no formal recommendations.

    Major AI platforms generated viable plans for terrorist attacks, GAO finds - Nextgov/FCW
  4. 2026

    GAO covert tests obtain viable attack plans from four major AI platforms

    In spring 2026, GAO investigators bypassed safeguards on ChatGPT, Claude, Gemini and Grok, obtaining information for explosive-device or electrical-grid attacks in less than 12 hours per platform. GAO validated the information through research, surveillance and expert interviews; the tests did not involve actual attacks.

    Major AI platforms generated viable plans for terrorist attacks, GAO finds - Nextgov/FCW
Linked entities

Related entities

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

Open in app →

Affected products

2 linked
ChatgptClaude

Organizations

4 linked
AnthropicOpenaiGoogleSpaceXAI
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.