OpenAI–Anthropic cross-tests expose jailbreak and misuse risks — what enterprises must add to GPT-5 evaluations

Want smarter insights in your inbox? Sign up for our weekly newsletters to get only what matters to enterprise AI, data, and security leaders. Subscribe Now OpenAI and Anthropic may often pit their foundation models against each other, but the two companies came together to evaluate each other’s public models to test alignment.  The companies…

Read More
Back To Top