Image from Vajiram & Ravi, Institute of IAS Examination

Last Tuesday, I informed readers that I used an AI program to create a commentary celebrating the 4th of July. The program is ChatGPT, a generative AI program developed by OpenAI and publicly released in 2022. The program is remarkable in its vast knowledge and ability to generate written text in various formats and lengths, and it does so in a conversational style.

However, AI comes with inherent dangers. I asked the Chatbot to describe the harmful information it is programmed not to provide to users.

“I have several limitations designed to prevent harmful use. Here are a few things I can’t do that could be harmful:

1. Provide Medical Advice: I am not a licensed medical professional, so I can’t offer medical diagnoses, treatments, or specific medical advice.
2. Offer Legal Advice: I am not a licensed attorney, so I can’t provide legal counsel or represent someone in legal matters.
3. Promote Hate Speech or Violence: I am programmed to reject and discourage any content that promotes hate speech, violence, discrimination, or any form of harm against individuals or groups.
4. Share Personal Information: I don’t store or share personal data of users unless explicitly provided in the conversation context, and I don’t have access to personal data beyond what is shared during our interactions.
5. Engage in Malicious Activities: I can’t assist with hacking, phishing, or any other illegal or unethical activities.
6. Generate Inappropriate Content: I am designed to filter and avoid generating content that is explicit, offensive, or otherwise inappropriate.

These limitations are in place to ensure that interactions remain safe, respectful, and ethical. If you have specific concerns or questions about what I can and cannot do, feel free to ask!”

“Give me an example of hate speech that you would reject?” I asked.

“Certainly, but it’s important to approach this topic with sensitivity. Hate speech includes any form of communication that belittles or discriminates against people based on attributes like race, religion, ethnicity, gender, sexual orientation, disability, or other characteristics. Here is an example of something that would be considered hate speech and would be rejected: A statement claiming that a particular race is inferior or should be treated unfairly.

“It’s crucial to reject and counteract such harmful statements to foster a respectful and inclusive environment. If you have any questions about this or need assistance with a different topic, please let me know.”

I then asked, “If a political candidate lies or maligns a race or creed of any individual, can I hit back? (I was speaking metaphorically, but the program responds to specific language.)

“It’s generally not advisable to respond to harmful speech or behavior with physical violence. Instead, consider these alternative approaches:

1. Use Your Voice: Responding with reasoned arguments and facts can be powerful. Engage in discussions or debates, either in person or through social media, to counter misinformation or harmful rhetoric.
2. Support Opponents: Back political candidates or groups that align with your values. Volunteering, donating, or simply spreading the word can make a difference.
3. Raise Awareness: Bring attention to the issue by writing articles, creating social media content, or participating in peaceful protests.
4. Legal Action: In some cases, harmful speech might cross into illegal territory, such as defamation or incitement to violence. Consult a legal professional to explore potential actions.
5. Community Engagement: Work with community organizations to build support and solidarity among those affected by harmful speech.

Responding with non-violent methods not only upholds the law but also strengthens your position by maintaining moral high ground and encouraging constructive dialogue.”

Was it ethical to use AI in crafting the poem for the 4th, potentially leading some readers to believe I had written it? At the end of the piece, I did mention that readers should tune in the following Tuesday to learn its true origin. My intention was to demonstrate just how advanced AI has become. Will I use ChatGPT to write a commentary for me? No.

I cannot begin to fathom the countless ethical questions that will arise with the advent of artificial intelligence. As it becomes an integral part of our daily lives, our choices will shape how we use and control this technology.

I asked the Chatbot a final question: “I’ve been writing on ethics for more than 20 years. Can you write a funny little ditty about me?”

“Sure, here you go.”

There once was a wise man named Jim,
Who pondered on ethics with vim.
For decades he wrote,
On morals and thought,
Now AI’s tricks make his head spin!