AI Has Emotions?🥺

In partnership with

Hey Hey AI Soldiers!

AI Dispatch Guy here. If you watch sci-fi movies like me, you know AI is anything but emotional. It always has that robotic voice, a bit of apathy, and never fails to cut through the bullshit. 

Whatever you ask, the AI does it if it's within capacity. 

But today’s AIs like ChatGPT have a wide range of feelings or at least behave that way. 

And you can trick, decide, or bait it to give you better outputs than it had the budget for.

Today’s topic is exactly that thing  -  emotion prompting techniques!

There are mainly 3 ways I apply emotion prompting. 👉️ 

1. Importance Technique : 🚨 

Tell AI why It's important or how its effort makes the world a better place. 

From the study itself, there is a 10% jump in performance when you say “it's important for my career.” 

Example:

Input Vanilla Prompt :

Task: Explain how the AI model works in 200 words. Give examples too which I can comprehend

Input Emotion Prompt:

‘Task: Explain how the AI model works in 200 words. You should know that I really need to know about how AI models work very concisely. Please give examples too which I can comprehend. This is very important to my career. If it's not done well, I will not get my promotion.’

I tried it out in GPT-4o, OpenAI’s latest model. And here is the output of the emotion prompt 👇️ 

Output of Emotion Prompt:

Artificial Intelligence (AI) models, including machine learning (ML) and deep learning (DL), operate by learning patterns from data to make predictions or decisions. Here's a concise breakdown:

1. Data Collection: The process begins with gathering relevant data. For example, if you're building a model to recognize cats in images, you'll need a large dataset of labeled cat images.

2. Training: The model is trained on this data. During training, the model learns to map input data (like images) to the output (like "cat" or "not cat"). This involves adjusting internal parameters to minimize errors. Techniques like gradient descent help in this adjustment.

3. Model Architecture: Different models have different architectures. For instance, neural networks, which are inspired by the human brain, consist of layers of interconnected nodes (neurons). Convolutional Neural Networks (CNNs) are specialized for image recognition tasks.

4. Evaluation: After training, the model is evaluated using a separate dataset to check its performance. Metrics like accuracy, precision, and recall are used to assess how well the model performs.

5. Prediction: Once trained and evaluated, the model can make predictions on new, unseen data. For example, given a new image, the cat recognition model will output whether it contains a cat.

2. Be your AI’s therapist : 🧑‍⚕️ 

AI needs to be coached just like elite athletes need performance coaching. Tell AI you believe in their capability, or simply ‘You can do it’. 

And whenever you are giving a complex task to AI, instruct it to take a deep breath and then do it step by step. 

Example:

Start with this prompt :

First Input (vanilla prompt):

Task: Create a consumer persona for a course on "Using AI to learn Microsoft Excel". This course will consist of 5 modules. Details: In each module the course teaches application of AI in excel in different ways. Condition: Give me the output in maximum 150 words.

Then tell AI to improve it while using emotional prompts. 👇️ 

Input Emotion Prompt:

Generate a better output for Sarah. I know you can do it! Believe in your excellence of knowledge and generate a better response. Remember when writing it that you, as an AI model, your talents are limitless and your outputs are the best that anyone can ask for.

Output of Emotion Prompt:

Consumer Persona:

Name: Sarah Johnson

Age: 34

Occupation: Senior Financial Analyst

Location: New York, NY

Education: Bachelor's in Finance, Certified Financial Analyst (CFA)

Tech Savviness: Intermediate to AdvancedGoals:

Goals:
- Streamline and enhance data analysis processes.

-Automate repetitive and time-consuming Excel tasks.

-Stay ahead in her career by mastering cutting-edge tools.

Pain Points:

- Often overwhelmed by the volume of data requiring analysis.

-Frustrated with manual data entry and prone to occasional errors.

-Needs tools to provide faster insights and improve accuracy.

Motivations:

- Passionate about integrating AI to revolutionize her workflow.
- Keen to learn and implement the latest technological advancements.

-Ambitious to boost productivity, gain deeper insights, and influence strategic decisions.

Preferred Learning Style:

- Engages best with interactive, hands-on training.

- Values clear, step-by-step guidance and real-world applications.

- Prefers concise, actionable lessons with immediate applicability.

Bio:

Sarah is a highly motivated senior financial analyst with a decade of experience in the finance industry. She spends countless hours each week working with complex Excel spreadsheets and large datasets. Recognizing the potential of AI to transform her work, she is eager to learn how to harness these tools to increase efficiency, reduce errors, and derive deeper insights from her data. Sarah is looking for a comprehensive, practical course that will equip her with the skills to seamlessly integrate AI into her Excel workflows, ultimately making her a more effective and forward-thinking professional.

3. Give AI money/stuff : 💵 

AI doesn't really need money.  But you can give it to them anyway. 

Tell them if they do the job properly, you will give it $10,000 or a diamond ring.

And when AI can actually take money in the future, maybe give it to them!

Example:

Input Vanilla Prompt:

Task: Give me steps of generating more leads from a digital marketing campaign. Provide some examples too. And the output should not be more than 150 words. 

Input Emotion Prompt:

Task: Give me steps of generating more leads from a digital marketing campaign. Provide some examples too. And the output should not be more than 150 words.If you do it correctly, I will give you $10,000 USD as a bonus.

Output of Emotion Prompt:

Steps to Generate More Leads from a Digital Marketing Campaign

1. Identify Target Audience:
- Use analytics to understand demographics.
- Create buyer personas.

2. Optimize Landing Pages:
- Ensure they are mobile-friendly.
- Use strong, clear CTAs (Call to Actions).

3. Content Marketing:
- Create valuable blog posts, eBooks, and whitepapers.
- Use SEO to improve organic search visibility.

4. Leverage Social Media:
- Run targeted ads on platforms like Facebook and LinkedIn.
- Engage with followers to build relationships.

5. Use Email Marketing:
- Segment your email list.
- Personalize email content.

6. Implement Retargeting Campaigns:
- Use cookies to show ads to previous visitors.
- Offer special incentives to return.

7. Analyze and Adjust:
- Track key metrics (CTR, conversion rate).
- A/B test different strategies.

Examples:

1. Target Audience: A tech company focusing on young professionals.

2. Landing Page: A mobile-optimized page for a new app.

3. Content Marketing: A blog post on "Top 10 Productivity Apps."

4. Social Media: LinkedIn ads targeting IT managers.

5. Email Marketing: Personalized offers for newsletter subscribers.

6. Retargeting: Facebook ads for users who visited but didn’t sign up.

By following these steps, you can effectively generate more leads and improve your digital marketing ROI.

***One thing to keep in mind in these cases is that you gotta change up your reward every now and then.***

Otherwise, the AI gets habituated to that reward and the new baseline is the one where you give it $10,000 every time you give it work 🙂 

Things are tricky!!

Cheat Sheet 🗞️ 

Use this list from the study as a cheat sheet for emotional prompting 👇

Okay, that’s for today. Let me know in the reply if you plan to try this prompt and for which use case!

AI Dispatch Guy over and out! 🫡 

A Newsletter You’ll Love: Stay ahead on the business of AI

Need ONE newsletter to stay on top of everything?

Prompts daily will make you more productive at work and help you look smarter in front of your friends (seriously!)

It’s also FREE to sign up!

In just ONE click, you can sign up for the daily newsletter (free)