GPT-Red detects indirect prompt injections with 84 percent success rate — significantly higher than the 13 percent achieved by human experts — and contributes to developing safer model generations.
OpenAI uses a specialized red-teaming model called GPT-Red to systematically identify and remediate prompt-injection vulnerabilities in new model versions.
Loop Engineering automates the control of AI agents through loops rather than manual prompt engineering and is seen by developers at Anthropic, OpenAI and Google as a new standard.
OpenAI GPT-5.6 Sol, Terra, and Luna are available on Amazon Bedrock, covering requirements from complex reasoning to cost-effective high-volume inference.
Deutsche Telekom integrates OpenAI models into customer service, internal processes, and network operations as part of its AI-native telecommunications strategy.
OpenAI launches GPT-5.6 and ChatGPT Work as an automated agent solution with integration into standard office tools, positioning itself against competitors like Anthropic.
