Prompt Injection Testing for AI Agents: OpenAI’s GPT-Red
OpenAI disclosed details of GPT-Red, an internal system to find prompt-injection flaws at scale, with reported gains in GPT-5.6 Sol defenses.
Access our complete library of market analysis, software updates, and tech trends.
Showing 6 of 23 insights
OpenAI disclosed details of GPT-Red, an internal system to find prompt-injection flaws at scale, with reported gains in GPT-5.6 Sol defenses.
TechCrunch and Forbes report Reflection lined up $1 billion in Nebius compute, a sign AI teams may need multi-year capacity planning.
TechCrunch reports Anthropic and Blackstone's $1.5 billion Ode launch signals growing enterprise focus on AI deployment services.
OpenAI says Codex and ChatGPT Work reached 8 million combined users. What the figure does and doesn't prove — and what teams should confirm before adopting.
Cribl said its CardinalOps deal adds AI-native detection engineering to improve threat coverage, lower data costs, and support SIEM changes.
Satya Nadella says enterprises pay for AI in cash and in proprietary knowledge. What the argument means for Copilot and Azure AI buyers — and what Microsoft gains from it.
Showing 1 to 6 of 23 results