That can occur in a cloud-based, closed-weight model as well. So this argument makes no sense.
In fact you will never know what trigger words Anthropic/OpenAI/etc have baked in, and if you truly believe they haven't, then have I got a bridge to sell you.
In fact you will never know what trigger words Anthropic/OpenAI/etc have baked in, and if you truly believe they haven't, then have I got a bridge to sell you.