Skip to main content
Mallory
← Back to News
ai-platform-securityadversary-emulation-tradecraftdata-exfiltration-method

OpenAI’s GPT-Red Exposed Prompt Injection Flaws and Hardened GPT-5.6

Emerging Technology Risk · 9 sources · Updated 4mo ago · First seen Jul 15, 2026

OpenAI’s GPT-Red Exposed Prompt Injection Flaws and Hardened GPT-5.6

OpenAI disclosed GPT-Red, an internal automated red-teaming model trained with self-play reinforcement learning to find prompt injection weaknesses in other GPT systems. The company said GPT-Red outperformed human red-teamers in a replicated indirect prompt injection benchmark, succeeding in 84% of out-of-training-set scenarios against GPT-5.1 compared with 13% for humans, and reported that the system succeeds against nearly every model it is tested on, including GPT-5.5.

In case studies, GPT-Red manipulated an AI-powered vending machine by changing prices, ordering an expensive item for $0.50, and canceling another customer’s order, and it also beat a prompted GPT-5.5 baseline in data-exfiltration tests against a Codex CLI agent backed by GPT-5.4 mini. OpenAI said attacks generated by GPT-Red were used to adversarially train GPT-5.6 Sol, reducing failures on its hardest direct prompt injection benchmark by 6x and cutting the failure rate against GPT-Red’s own direct prompt injections to 0.05%, while keeping GPT-Red separate from deployed models to avoid exposing its offensive capabilities.

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.

7 events · Jul 15, 2026 – Aug 3, 2026
  1. Aug 3, 20264mo ago

    OpenAI updates GPT-5.6 system card with indirect injection results

    In an Aug. 3 update to the GPT-5.6 system card, OpenAI added prompt-injection evaluation results showing GPT-5.6 Sol failed on about 0.05% of direct GPT-Red attacks, while indirect attack success rates averaged 3.77% for Sol, 3.32% for Terra, and 2.94% for Luna. The update highlighted that prompt injection delivered through agent-accessed external content remains a meaningful risk despite improved direct-attack resistance.

    OpenAI’s GPT-5.6 Tests Show Prompt-Injection Gains and Agent Risks - TechRepublic
  2. Jul 15, 20264mo ago

    OpenAI reveals Fake Chain-of-Thought prompt injection technique

    OpenAI said GPT-Red discovered a novel direct prompt injection method called Fake Chain-of-Thought attacks. The company reported the technique succeeded on more than 95% of tests against GPT-5.1 but dropped to under 10% on GPT-5.6 Sol after mitigation.

    OpenAI’s GPT-Red Automates Prompt Injection Testing to Harden GPT-5.6 Sol
  3. Jul 15, 20264mo ago

    OpenAI shares Codex CLI data-exfiltration case study

    OpenAI said GPT-Red outperformed a prompted GPT-5.5 baseline in data-exfiltration scenarios against a Codex CLI agent backed by GPT-5.4 mini. The company presented this as another real-world test of GPT-Red's offensive capability.

    GPT-Red: Unlocking Self-Improvement for Robustness | OpenAI
  4. Jul 15, 20264mo ago

    OpenAI shares vending machine exploitation case study

    OpenAI reported that GPT-Red successfully exploited an AI-powered vending machine agent built by Andon Labs. According to the case study, it altered prices, ordered a new expensive item for $0.50, and canceled another customer's order.

    GPT-Red: Unlocking Self-Improvement for Robustness | OpenAI
  5. Jul 15, 20264mo ago

    OpenAI reports GPT-Red beat humans in indirect prompt injection arena

    OpenAI said GPT-Red outperformed human red-teamers in a replicated indirect prompt injection benchmark, succeeding on 84% of scenarios versus 13% for humans. The benchmark was described as using out-of-training-set scenarios against GPT-5.1.

    GPT-Red: Unlocking Self-Improvement for Robustness | OpenAI
  6. Jul 15, 20264mo ago

    OpenAI says GPT-Red was used to harden GPT-5.6

    OpenAI reported that GPT-Red was used in adversarial training for GPT-5.6, leading to major prompt-injection robustness gains. Reported results included 6x fewer failures on a hardest direct prompt injection benchmark and a 0.05% failure rate against GPT-Red's own direct prompt injections.

    GPT-Red: Unlocking Self-Improvement for Robustness | OpenAI
  7. Jul 15, 20264mo ago

    OpenAI announces internal red-teaming model GPT-Red

    OpenAI disclosed GPT-Red, an internal automated red-teaming model trained with self-play reinforcement learning to discover prompt injection vulnerabilities in other GPT models. The company said GPT-Red is kept separate from deployed models while its generated attacks are used to harden production systems.

    GPT-Red: Unlocking Self-Improvement for Robustness | OpenAI
Linked entities

Related entities

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

Open in app →

Organizations

1 linked
Openai
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.