To address the issue of the AI-based fraud detection system flagging genuine transactions as fraudulent due to data poisoning, the primary focus should be on:
* Retraining the AI Model:
* Verified Data: Use only verified, clean transaction data to retrain the model. This helps to eliminate any compromised data that might be skewing the AI's decision-making process.
* Model Integrity: Ensure the integrity of the training data to prevent future data poisoning attempts.
* Real-Time Monitoring:
* Detection Systems: Implement real-time monitoring to detect any attempts at data poisoning as they happen. This involves setting up alerts for abnormal patterns that could indicate malicious
* activity.
* Continuous Learning: Integrate continuous learning systems that can adapt and respond to new threats in real-time, ensuring the AI system remains robust against evolving attack vectors.
References:
* NIST guidelines on AI and data integrity:NIST AI
* Research on data poisoning and mitigation techniques:IEEE Xplore

The correct answer is A, as it identifies the type of Internet access policy implemented by Ashton in the above scenario. An Internet access policy is a set of rules and guidelines that defines how an organization's employees or members can use the Internet and what types of websites or services they can access. There are different types of Internet access policies, such as:
* Paranoid policy: This type of policy forbids everything and imposes strict restrictions on all company computers, whether it is system or network usage. This policy is suitable for organizations that deal with highly sensitive or classified information and have a high level of security and compliance requirements.
* Prudent policy: This type of policy allows some things and blocks others and imposes moderate restrictions on company computers, depending on the role and responsibility of the user. This policy is suitable for organizations that deal with confidential or proprietary information and have a medium level of security and compliance requirements.
* Permissive policy: This type of policy allows most things and blocks few and imposes minimal restrictions on company computers, as long as the user does not violate any laws or regulations. This policy is suitable for organizations that deal with public or general information and have a low level of security and compliance requirements.
* Promiscuous policy: This type of policy allows everything and blocks nothing and imposes no restrictions on company computers, regardless of the user's role or responsibility. This policy is suitable for organizations that have no security or compliance requirements and trust their employees or members to use the Internet responsibly.
In the above scenario, Ashton implemented a paranoid policy that forbids everything and imposes strict restrictions on all company computers, whether it is system or network usage. Option B is incorrect, as it does not identify the type of Internet access policy implemented by Ashton in the above scenario. A prudent policy allows some things and blocks others and imposes moderate restrictions on company computers, depending on the role and responsibility of the user. In the above scenario, Ashton did not implement a prudent policy, but a paranoid policy. Option C is incorrect, as it does not identify the type of Internet access policy implemented by Ashton in the above scenario. A permissive policy allows most things and blocks few and imposes minimal restrictions on company computers, as long as the user does not violate any laws or regulations. In the above scenario, Ashton did not implement a permissive policy, but a paranoid policy. Option D is incorrect, as it does not identify the type of Internet access policy implemented by Ashton in the above scenario. A promiscuous policy allows everything and blocks nothing and imposes no restrictions on company computers, regardless of the user's role or responsibility. In the above scenario, Ashton did not implement a promiscuous policy, but a paranoid policy.
References: , Section 6.2