Skip to content

MartinYeung5/AIStarknet-anti-fraud

Folders and files

NameName
Last commit message
Last commit date

Latest commit

44ddf85 · Dec 12, 2024

History

58 Commits
Oct 25, 2024
Nov 23, 2024
Dec 12, 2024

Repository files navigation

AIStarknet-anti-fraud

Image

alt text

About this project

Nowadays, we are facing different frauds in our life. Especially in the rapid changing generation, we may use different online applications such as online shopping, gaming, online services, and offline applications such as interaction with other people or organization. During these activities, we may suffer from lots of fraud risks and we should find some ways to protect our benefit and prevent to become victim. However, most of people may not have enough knowledge or experience to prevent some frauds. Therefore, this project can help them to easily discover the potential risk of fraud with innovative technology.

Tagline

Save time to use our product to protect your benefit in more comprehensive ways

The problem it solves

  1. most of people may not have enough knowledge or experience to prevent some frauds.
  2. no one can provide immediately analysis on specific information and let you know whether you are facing the frauds or not.

With this project, it can provide a solution:

  • help user discover the fraud risk of different information such as platform, event, and other action.
  • send fraud alret when discover potential risk of fraud based on user life record (the record will be provided or collected from user).

Frauds example

Example and some tips for prevention (3 mins video):

Phishing Websites

  • Set up a fraudulent website with the same style as the genuine website and install a malware into the fraudulent website to obtain the login information of the targets or members of the target organisation

Phishing Emails

  • Send emails with attachments of hidden malicious programs or fraudulent website links to infect the target computer network

Online investment Fraud

  • Through online social media platform, forums or instant messengers, fraudsters promote investments with low risks and high returns and attract netizens to invest. However, most of the investment plans do not exist or are filled with traps. Some fraudsters falsify transaction records that appears to be gaining profits, luring victims into investing more money. Eventually, fraudsters can no longer be reached.

Credit Card Fraud

  • Once credit card information including card number, expiry date and CVC falls into the hands of criminals, your credit card information may be misappropriated and you may suffer loss.

Web3 Fraud

  1. Fake application: it will be risked. when someone downloads the fake web3 application, they may suffer from the high risk of fraud.
  2. Complicated signature request The request may include diffrent actions which can get your asset or gain more access rights and so on. alt text

Identity Fraud

  • AI is a double-edged sword, it can explore the dual role of artificial intelligence in reshaping fraud prevention and providing new opportunities for fraudsters. Deepfake fraud are growing at an unprecedented rate year-on-year, with China recording a 28 times increase, South Korea approximately 16.3 times (1625%), Turkey approximately 15.3 times (1533%), Singapore 11 times, and Hong Kong 10 times. alt text

Summary

Actually, it still has different types of fraud. Therefore, we may need to pay more attention to prevent these frauds.

Data reference

  1. identity fraud report

Target market

Hong Kong

In daily life, we may have risk to encounter various frauds. However, most of the protection may not care your life's risk. Those software/platform/tool may just tell you general risk and can't tell you about the risk in your real life. Furthermore, different cities may have various popular frauds or specific frauds based on the local culture/lifestyle/goverment policies. Therefore, it is important to develop an application to protect your benefit more effectively.

Research

  1. Technology Crime Figures alt text alt text

  2. Online Investment Fraud alt text

  3. Phishing Scam alt text

  4. E-shopping fraud alt text

  5. Credit card misuse alt text

  6. Email scam alt text

  7. Theft of online account alt text

Fraud statistic (Hong Kong only)

A total of 19 897 cases of fraud were registered in the first half of 2024, an increase of 1 154 cases (+6.2%) compared with the same period last year (18 743 cases), of which 62.3% were Internet-related. Fraud cases accounted for 43.9% of the total number of crimes and the amount involved was $4.48 billion (+1.79 billion, +66.5%). The loss incurred from investment fraud was $1.55 billion, accounting for 34.5% of the total amount involved.

The report from TransUnion found that 6.6% of all digital transactions where the consumer was in Hong Kong were suspected to be Digital Fraud in 2023. This number is 32% higher than the global suspected Digital Fraud rate of 5% over the same time period.

Reference:

  1. Hong Kong Government (2024)
  1. TransUnion (2024)

Current applications (competitors)

According to the reserach, lots of "similar" applications in the market. However, they may focus on different specific market or pain points. such as only focus on web3 world, email section, web link. you can image if you want to get better protection, you may need to download lots of useful applications. Also you may need to learn how to use and may need to pay the extra fee. However, for the general user, they may only a few applications and don't want to spend so much time on checking/learning. Moreover, they may care more about the protection on their daily life. Therefore, most of the applications may not fit into their needs.

Key features

  1. Analyze the fraud's risk of information
  2. Generate fraud alert to the user when they are using the service
  3. Report the fraud case when they encountered
  4. Provide more support on your daily life not only on Web3, but also in offline society
  5. Collect the fraud report from different organizations and platforms and social media groups (such as facebook group, TG group, forum group) by AI

What is the product

  1. web platform
  • user can check the fraud risk by specific data input including email address, link, web3 wallet address, phone number and so on.
  • user can import their life data inlcuidng phone number list, email address list, schedule data for checking.
  • user can report the fraud case

Project structure

1. Starknet

  • 1.1. smart contract
  • using account abstraction (Can provide the fraud monitoring)

2. Web frontend

  • 2.1. main page design - including basic information of project
  • 2.2. user dashboard - after user login, user can do different actions in dashboard
  • 2.3. whole process of audit contract
  • 2.4. audit report (security analysis)

3. AI

  • 3.1. using Gemini to preform the analysis process of fraud risk: alt text alt text

  • "contract audit" is developed (using gemini)

  • user can submit their contract, then system will analyze the contract and generate the audit report in order to let user to know what is the risk and risk level of the contract. Therefore, it can protect user benefit.

4. function development

  • user login (google login)
  • audit contract function
  • subscription fee function (payment by credit card)

What's next for

  1. Tool/plugin for web3 wallet
  • It aims to help user can use our serveic anywhere

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages