Anthropic just shut OpenAI out of its Claude API after catching them using Claude Code to test GPT-5. Is this a justified move to protect proprietary tech, or is benchmarking a necessary part of AI progress? As AI models get more advanced, where should we draw the line between healthy competition and crossing ethical boundaries? Would you want your favorite tools locked down, or is open benchmarking better for innovation? #Tech #ClaudeAPI #OpenAI