Please enable JavaScript.
Coggle requires JavaScript to display documents.
AI and ML - Coggle Diagram
AI and ML
Cybersecurity Risks
Adversarial Attacks
AI systems can be manipulated by feeding them malicious data, creating vulnerabilities.
Bias & Discrimination
AI may introduce biases in security decisions, targeting specific groups unfairly.
AI and ML systems can exhibit bias or produce unintended consequences due to biases present in the training data or the algorithms themselves
-
Privacy Concerns
AI and ML require vast amounts of data for training and operation, often involving sensitive or private information.
-
Scenario
Scenario 1: A hacker uses AI to create a deepfake of a company executive’s voice, deceiving employees into transferring company funds, leading to financial losses.
Ethical Option
Implement stronger authentication measures that go beyond voice verification, such as multi-factor authentication (MFA), to ensure no fraudulent activity occurs.
Unethical Option:
The company could neglect security updates, relying solely on outdated voice recognition systems, making it easier for hackers to succeed.
Ethical Theory
Utilitarianism
-
By adopting stronger security measures, the company is preventing potential harm not only to its immediate finances but to the well-being of all stakeholders.
Virtue Ethics
-
A company that exhibits virtues like responsibility and foresight would naturally implement stronger security measures to prevent fraud
My Choice
Ethical Option
Protecting against fraud with stronger verification methods ensures the safety of company funds and upholds corporate responsibility. The company not only prevents potential fraud but also upholds its moral integrity and commitment to fostering trust, safety, and ethical leadership
The company not only fulfills its duty to protect its stakeholders but also maximizes the overall benefits by avoiding potential harms like financial loss and reputation damage. This decision aligns with long-term security and well-being for the greatest number of people.
Scenario 2: AI flags certain ethnic groups as higher cybersecurity risks due to historical data biases.
-
Unethical Option
Allow biased decisions to persist, targeting specific groups unjustly.
My Choice:
Ethical option.
AI must be audited and corrected to avoid discrimination and ensure fairness across all demographics.
From theory of Justice, AI should promote fairness, ensuring that all individuals are treated equally.
According to Virtue Ethics, it is essential to act with fairness and justice by auditing the AI system to remove biases and ensure equitable treatment across all demographic groups. This not only aligns with ethical conduct but fosters trust and long-term societal benefits.
-
Ethical Challenges
Bias and Discrimination
AI and ML systems learn from data, and if that data reflects existing societal biases, the resulting systems can perpetuate and even amplify those biases
Job Displacement
AI can automate tasks, leading to job loss, which poses an ethical dilemma for society.
-
-
Scenario
Scenario 1: A company uses AI to automate hiring decisions, leading to potential biases in selecting candidates.
-
Unethical Option
Allow AI to make final decisions, potentially reinforcing biases and unfair hiring practices.
My Choice
Ethical option.
-
According to utilitarianism, the ethical option of involving human oversight maximizes the overall well-being by ensuring fair hiring practices and preventing the negative impact of bias, thus benefiting both the candidates and the company in the long run.
Based on Kantian ethics, the ethical option is to include human oversight. This ensures that candidates are treated fairly and with respect, adhering to the principle that all individuals should be treated as ends, not just means for efficient hiring.
From a virtue ethics perspective, the ethical option involves human oversight, which reflects a company’s commitment to acting with fairness, wisdom, and integrity. By prioritizing fairness and ensuring just treatment for all candidates, the company aligns its hiring process with virtuous principles.
Scenario 2: A factory introduces AI-powered systems to automate tasks, leading to layoffs. Workers face unemployment, but the factory increases profits and productivity.
Ethical Option
The company offers retraining programs to displaced workers, helping them transition to new roles within or outside the company.
Unethical Option
The company proceeds with automation without offering any support to the affected workers, leaving them to face unemployment.
My Choice
Ethical Option
Providing retraining programs balances the company’s need for efficiency with the workers’ right to fair treatment. This approach promotes long-term community welfare.
From a utilitarian perspective, as retraining workers increases the net positive outcome by improving their future prospects and reducing societal harm.
From a Kantian Ethicsl perspective, the ethical option aligns with the idea of respecting workers’ inherent worth and promoting fairness.
From a virtue ethics perspective, acting with compassion and responsibility, the company should help workers transition and not simply prioritize profit over people’s livelihoods.
Cybersecurity Advantages
-
-
Scenarios
Scenario 1: A bank uses AI to monitor transactions for fraud and prevent cyber-attacks. It flags an unusually high withdrawal, freezing the customer's account.
Ethical Option
The bank can monitor transactions using AI but should ensure there is a quick and effective process for human intervention when legitimate customers are flagged. This way, customers can regain access to their accounts quickly
Unethical Option
Over-reliance on AI without human oversight causes undue distress to legitimate customers, creating more harm than benefit. The AI might wrongly flag legitimate transactions, freezing accounts for an extended period and negatively impacting the customer.
My Choice
Ethical option
-
-
Customer needs addressed
Quick, human intervention process for flagged legitimate customers.
Restores account access promptly, minimizing disruption.
Maximized satisfaction
-
System responds efficiently to individual needs, balancing security with convenience.
Ethical Theory
Virtue Ethics, ensure fairness and balance between security and customer experience.
Utilitarianism promotes the greatest good by ensuring a balance between security and customer experience.
-