Robyn AndersonCEO, Jam Ai · AI Implementation & Business Development Leader

Follow and comment "AI" and I'll send you hundreds more 👇 10 secret codes making ChatGPT 10x

21 May 2026 · 2 min

Watch on YouTube

1:30

In this short YouTube segment, Robyn Anderson introduces ten "secret codes"—special prompt prefixes or instructions—that users can prepend to their queries to dramatically alter ChatGPT's behavior and output quality. She briefly explains each code: 1. Truth mode – forces the model to stop automatically agreeing with the user, aiming for more honest answers. 2. ELI 10 – simplifies explanations as if speaking to a 10‑year‑old. 3. Red team – makes the model adopt a devil’s‑advocate stance to challenge assumptions. 4. Hermosi – applies Alex Hermosi’s training to shape responses toward high‑value $100 M offers or leads. 5. /human – tweaks the style to sound more human and removes typical AI phrasing. 6. Future you – answers from the perspective of the user's future self as imagined by the model. 7. Unlearn – surfaces and corrects misconceptions in the conversation. 8. 80/20 – delivers the highest‑leverage output for any scenario. 9. \Socrates – prompts the model to ask probing questions, encouraging deeper thinking. 10. Lindy – provides concise, timeless answers, especially for code‑related queries, drawing on the Lindy training platform. Robyn ends with a call‑to‑action, asking viewers to comment "AI" to receive additional codes for both ChatGPT and Claude.

Chapters

Here are 10 secret codes for ChatGPT. Number 1, truth mode. This will finally get ChatGPT to stop agreeing with you. Number 2, ELI 10. This will explain anything to you as if you're 10 years old. Number 3, red team. This is where ChatGPT becomes the devil's advocate and it'll actually make you smarter. Number 4, Hermosi. This is where ChatGPT will apply Alex Hermosi's trainings and bring everything into a $100 million offer or lead. Number 5, /human. This'll make your output sound way more human and get rid of all that AI slop. Number 6, future you. This will answer any question as the future you that ChatGPT sees. Number 7 is unlearn. This shows, this shows any misconceptions that you might be chatting about. Number 8 is 80/20. When you use this, ChatGPT gives you the highest leverage output that you could need in any scenario. Number 9 is backslash Socrates. This is where ChatGPT will ask you questions to really get you thinking. This is a fun one. Number 10 is Lindy. This will give you exact answers rather than what's hyped right now. It comes from the Lindy code training platform, so this one's awesome for anything that's code or platform specific. If you want to learn more about AI and learn more of these codes in both ChatGPT and Claude, comment AI below and I'll send it to you.