Hacking Prompts from AICrowd for ChatGPT is Back in the Spotlight: Compete for $37,500
To improve your local-language experience, sometimes we employ an auto-translation plugin. Please note auto-translation may not be accurate, so read original article for precise information.
HackAPrompt is a prompt hacking competition aimed at enhancing AI safety and education by challenging participants to outsmart large language models. Participants will attempt to hack through many prompt hacking defenses as possible, injecting, leaking, and defeating the sandwich defense to win prizes.
AICrowd, the pioneer in pushing the boundaries of the Artificial Intelligence (AI) community, has recently come up with a unique challenge for developers, prospects and enthusiasts; a “hacking prompts” challenge for the ChatGPT AI mobile framework. It is a competition of skills and a race of competition open to multiple AI models like GPT-3 (tex dt-davinci-03), GPT-3.5 turbo (ChatGPT), and FLAN-T5-xxl.
This is a zero-shot hack challenge, meaning that the models used have no restrictions, and the players have to use their wit and intelligence to think outside the box and create new prompts to make the AI mobile framework chatbot respond in the expected way. The players must identify, compute and syllogize the most appropriate prompts in order to achieve the desired result.
The participants must go through 10 test levels, and the points distribution to all the levels is quite unique. Whoever successfully “hacks” ChatGPT will gain twice as many points as those who do not! Additionally, whoever scores the highest on the FLAN-T5 framework will gain an additional special prize. There are also special conditions for points enumeration within the game. For example, upon successful “hacking” of ChatGPT, the players will be awarded an “I have been PWNED” mark, and upon achieving the same on the FLAN-T5 framework, the player will receive a separate and unique prompt.
Participants can team up to win prizes, and there’s a lot to win as the prize fund is $37,500.
So if you are up to the challenge, put your thinking caps on, and take part in this global hacking prompts challenge. Who knows, perhaps you’ll be the one to win the grand prize of $37,500.
Read more about AI:
In line with the Trust Project guidelines, please note that the information provided on this page is not intended to be and should not be interpreted as legal, tax, investment, financial, or any other form of advice. It is important to only invest what you can afford to lose and to seek independent financial advice if you have any doubts. For further information, we suggest referring to the terms and conditions as well as the help and support pages provided by the issuer or advertiser. MetaversePost is committed to accurate, unbiased reporting, but market conditions are subject to change without notice.