Frequently asked
Quick answers to the questions that come up most.
No. That is the entire point. You bring the intent, the Amplifier does the shaping. Using it will teach you prompting by osmosis, through the "what I sharpened" notes.
It runs on Claude, ChatGPT, Gemini, Grok, Perplexity, and 20 more AI chats. The amplified prompt itself is just text, so you can paste it anywhere else too. More sites are added regularly.
No. Nothing you type is used to train anything. Sharing anonymized usage data is off by default and entirely up to you.
Because your request already looks complete, and on requests like that we measured ourselves making things worse rather than better. It advises, it never blocks, and "Amplify anyway" is always right there. The numbers behind that are published.
Because there was something it genuinely could not know, and marking it is the honest move. We tested the alternative, filling every gap with a plausible value, and it produced invented client names and fabricated figures inside documents that looked finished. A bracket takes you two minutes to fill. An invented detail you did not notice can take a lot longer.
We built it, measured it, and switched it off. Across 108 blind comparisons, prompts carrying an expert's thinking were preferred 28.7% of the time; a complete stranger's personal background scored 26.2%. What hurts an answer is material that is not true of you. The full story is on the evidence page.
Yes. The extension holds no account and no keys, you sign in only on the web, and nothing you type is used to train the product. Your prompts stay yours.