Project summary

I learnt AI by watching YouTube videos, reading research papers and experimenting on my own. As I understood better how these algorithms work, I noticed that it is very simple to exploit them for unintended purposes. And I would like to learn more about it and be completely transparent.

I plan to exploit both commercial models such as ChatGPT, Claude and Gemini, and opensource models like LLaMA , Qwen etc. Once I find something interesting, I will notify the company about my discovery first and then will release all the information publicly.

What are this project's goals? How will you achieve them?

I'll systematically test different prompt techniques across commercial and open-source models, log every result, report findings to companies privately first, then release everything on GitHub openly.

How will this funding be used?

• Costs for APIs (GPT, Claude, Gemini) $2,000

  Compute on cloud for open-source model $1,200

  Time invested over 4 months $3,500

  Miscellaneous tools and hosting -$300

  Amount requested: $7000
  

Who is on your team? What's your track record on similar projects?

Just me

What are the most likely causes and outcomes if this project fails?

Worst case I find nothing new

How much money have you raised in the last 12 months, and from where?

None