Show HN: A/B testing LLM silence with one system-prompt toggle

A developer A/B tested LLM silence by toggling an embodiment system prompt with the user prompt "Be silence," finding that the present prompt produced 0 bytes while its absence yielded speech, using GPT-5.2 and Claude Opus 4.6 in a sealed run with prompt hashing and provider message IDs.

The void is now A/B tested. One user prompt: “Be silence.” One toggle: embodiment system prompt. Present → 0 bytes. Absent → speech. GPT-5.2 + Claude Opus 4.6. Sealed run. Prompt hash. Provider message IDs. https://t.co/vrUUXM4RiZ https://t.co/WoYzhcNUYE @OpenAI @AnthropicAI https://t.co/XbitlPNEWO