A bounty smart contract for trusted exam questions

I recently came across the idea of smart contract, where people basically stores a set of instructions in a blockchain, which will then be forcefully executed when some criteria were met.
After knowing this idea, I thought it would be interesting if I could play with this idea and come up with my own use-case for smart contract!

Background

How can we ensure an exam is fair?
One prerequisite for a fair exam is the questions are kept secret before the exam, and are only release to the student until the exam. And I thought, I can use the idea of smart contract to ensure a decentralised trust to ensure unleak exam questions.

Few works have been published relating cryptography and exam [1][2][3], which mainly focus on securing exam questions and avoid them leaking to students.

Unlike most prior blockchain-based exam schemes, they largely assume an honest exam authority. My construction here tries to reduce the trust placed in the professor by making exam leakage economically costly through a smart-contract bounty mechanism.

The Method

The professor make the following preperation:

1. The professor prepare a substential amount of cash in crypto currency. Say 1 bitcoin (ideally).
2. The professor setup a bank of questions (say 10000 questions), where each questions are indexed, this shall be possible with modern generative AI.
3. The professor selects a bag of questions from the bank (say 10 questions), which will be asked in the exam.
4. The set of selected questions along with a key are hashed, with the key required non-trival computational time to compute so that it meets an requirement (e.g. the key has to be the smallest 64bit integer such that when concatnated with the exam question index and hashed, the end of the hash has 8 zeros)

With the above prepared, the professor setup a smart contract on a blockchain a period of time before the exam. The contract state:

1. The mentioned substenial amount of cash is locked, until the questions are released, unless:
2. If someone provided valid combination of {exam question indices + key} to produce the exact hash, then the amount of cash will be sent to that person, and the {exam question indices + key} will be released, making the exam question known to everyone.

This effectively sets up a bounty for people to guess the exam questions.

Discussion

I would explain some of my thinking in this session.

Why a bank of questions and index from them?
This is to prevent the professor modify the question’s wordings a little bit, then leak it to someone. If this is what the professor does, then that person could still use the modified questions, guess the possible questions to be asked in the question bank, which greatly reduce the computational time needed to obtain the {exam question index + key}

The key that require non-trival time to compute
If the professor simply hash the exam question indices without the need for a key, one could simply run through all combination of indices to brute force the hash. By making the key non-trival to compute, this could greatly improve the security of the proposed method

Security against brute force
If the exam has $m$ questions, and the question bank has $N$ questions, the time complexity to brute force the contract will be $\Theta({N \choose m})$. For fixed $m$ against $N$(which shall be typical for exam question), the time complexity is bounded to be:

${N \choose m} = \frac{N(N-1)\cdot \cdot \cdot (N - m + 1)}{m!} \leq \frac{N^m}{m!} = O(N^m)$

For example, for an exam with 5 questions, with question bank size of 100, and each guess takes 0.1 second to compute, then the expected time needed to brute force the contract would be about 31 years. So the student should be better off studying for the exam instead of trying to brute force the bounty.

Conclusion

I hope this is interesting! Overall, this illustrate a use for smart contract without the need of real world information. A bounty like construction for trust would also be very interesting to explore more. Thank you for reading this post.

References

[1] Nagarajan, B., Ananth, C., & Mohananthini, N. (2023). Blockchain-based smart and secured scheme for question sharing using bee colony optimization based quantum logistic map encryption. International Journal of Information Technology, 15, 2889–2895. https://doi.org/10.1007/s41870-023-01333-4

[2] Pranathi, P., Yojith Siddeshwar, G., Laxmi, S., & Roja, G. (2025). Question paper leakage prevention using blockchain. International Journal of Multidisciplinary Research in Science, Engineering and Technology, 8(5), 8385–8390. https://doi.org/10.15680/IJMRSET.2025.0805101

[3] Tejashwini, Y., Farzanula, M., Nikhil, S., & Khan, S. A. (2025). Question paper leakage protection using blockchain. International Journal of Innovative Research in Electrical, Electronics, Instrumentation and Control Engineering, 13(9), 19–27. https://doi.org/10.17148/IJIREEICE.2025.13904

If you are interested in citing this post:

@article{hon2025bounty,
  author = {Anson Hon},
  title = {A bounty smart contract for trusted exam questions},
  year = {2025},
  month = {November},
  note = {Unpublished manuscript},
}