In today’s column, I am taking a modest shift from covering hard-hitting topics about generative AI and will be focusing on the seemingly light-hearted matter of identifying helpful uses of generative AI for boosting your Valentine’s Day splendor. That being the case, don’t worry, I am nonetheless going to sneak into this heartfelt discussion a slew of important insights about the strengths and weaknesses of generative AI.
Yes, this is one of those instances of using some sugarcoating to get across what might be the best of times and the worst of times about the nature of modern-day generative AI.
Insights About Generative AI Via The Revered Valentine’s Day
We all have been inundated with the wonders of generative AI. Daily headlines exhort that generative AI is the best thing since sliced bread. On the one hand, there are certainly crucial capabilities that the latest in generative AI offers to us. From work to play, you can readily make use of generative AI in quite substantial ways. No doubt about it, see my extensive analyses at the link here.
At the same time, it is equally vital to acknowledge the limitations of today’s generative AI. Let’s put a damper on the outsized outstretched hype. Let’s keep our feet on the ground and soberly consider what generative AI can attain. I thought that perhaps by going through various examples of the use of generative AI for Valentine’s Day purposes, we can agreeably explore the benefits and detriments of contemporary generative AI.
In that sense, you are about to read a twofer.
You can luxuriate in learning about how to boost your Valentine’s Day endeavors via leveraging generative AI. Put one in the plus column. Meanwhile, without even necessarily realizing it is happening, you’ll get a chance to plainly see what generative AI can do and what it cannot do. That is the twofer angle. Two key elements dovetailed into one engaging discussion. Score two resounding points.
MORE FOR YOU
I sincerely hope that in some ample fashion, this helps your Valentine's Day considerations and that you also feel a tad more informed about generative AI by the time you finish reading this edifying analysis. Furthermore, there is a practicality on top of this, namely that I am going to use the widely and wildly popular generative AI app known as ChatGPT to showcase the actual real uses of generative AI in this context. If you’ve never used generative AI, the examples will give you a tangible sense of what it is like.
For those of you who are using generative AI and if you want to try the same prompts that I show here, you are welcome to do so, though a caveat is worthwhile noting.
The caveat is this.
Generative AI is a mathematical and computational pattern-matching formulation that relies on probability and statistics. As such, any given prompt will (perhaps surprisingly to some) generate a different response each time it is used. This is because the generated wording is being assembled by the generative AI based on probabilities. The goal of the AI developers is to have generative AI that devises sentences and responses that seem to be unique or original. By changing up whatever words are being laid out, the responses have a human-like appearance as to being spontaneously stated.
Thus, you could proceed to enter the same prompts that I show in the examples here and yet get somewhat different responses from ChatGPT. That’s fine. The odds are that the essence of the replies will be nearly the same. It would be rarer that a given stated prompt would get an extremely different response. I am not saying that is not possible. I am only saying that due to the pattern-matching formulations, the chances are that you will get quite similar replies that vary only in minor wording choices.
By the way, did you see what just happened there?
I managed to get a morsel of insights about how generative AI works and did so without having to raise sharp flags or set off loud horns. We can glean useful facets about generative AI in the course of normal discourse. I will do this as we proceed into the Valentine’s Day facets.
For those of you who might also desire ongoing hard-hitting straight-out coverage of the latest in AI trends, see my column at the link here.
Your Seven Handy Ways To Use Generative AI For Valentine’s Day Preparations
Valentine’s Day is this coming Wednesday, February 14.
That shouldn’t be a shocker.
Of course, if you didn’t realize that this heralded magical day was fast approaching, you’d better get your game on and get going. Loved ones who are silently expecting you to do something for them on Wednesday are going to be disappointed or maybe steamed if you are essentially a no-show or have zero to share with them.
The good news is that we can use generative AI to rescue you from landing in the doghouse.
I was going to provide about a half-dozen examples of using generative AI for Valentine’s Day planning and readiness, but then I realized that since Valentine’s Day is on the fourteen of the month, and since seven is half of fourteen and that seven is my lucky number, I might as well round things out with seven examples.
The seven examples are primarily about getting ready for Valentine’s Day. I also snuck into the list a few aspects that are probably handy mainly after Valentine’s Day. Here’s why. You might want to come back to generative AI post-celebration and see what else the AI might help you with on a go-forward basis. Maybe to recover from a bruised soul, though I hope that’s not the case.
I suppose you could also use generative AI in real time on Valentine’s Day.
Let me briefly say something about that.
Not everyone is necessarily a fan of generative AI. Some believe that the use of generative AI is a form of cheating in all manner of ways. For example, if you use generative AI to give you some ideas of what to say to a loved one, the chances are that the words you then utter will be construed as hollow and faked. You are supposed to be speaking from the heart. Using an emotionlessly cold and calculating AI to inform you about what to say is seen as a falsity and a travesty.
What’s kind of interesting about that strict viewpoint is that if you went online to the Internet and did a search for ideas or suggestions about what to say to a loved one, few would seem to be so quick to condemn you. They might even compliment you on being caring enough to do some handy background research. Yet, using generative AI for a similar purpose gets their goat.
I would certainly suggest that you not word-for-word repeat whatever generative AI has indicated to you. That seems overboard. Using generative AI to spur your thoughts and formulate your own words of love seems to me to be a fair and balanced form of usage. This reasonably smacks of reading a prized romance novel or seeking out love note suggestions on the Internet.
The zillion-dollar question is whether you should reveal to the other party that you essentially consulted generative AI for Valentine’s Day.
That is quite a conundrum.
If you say that you did consult generative AI, doing so might ruin the entire occasion. All the fresh and fluffy air might be taken right out of the veritable Valentine’s Day balloon. The other person might react negatively. They might be greatly disappointed and hurt to their core.
If you don’t say that you consulted with generative AI, but if they find out you did, now you seem to be in double trouble. You used generative AI, which is one strike potentially against you. And, perhaps as an even greater crime, as it were, you omitted saying that you did so. You have committed an offense by omission. You lied by not being forthright. Maybe you can shrug it off by claiming that you didn’t see any reason to mention that you had used generative AI. It seemed like a trivial point and unworthy of bringing up.
Wow, a mind-bending dilemma.
The other possibility is that you do state that you used generative AI and the other person is delighted to know so (smiley face scenario). They say to you that it was especially thoughtful of you to use modern AI on their behalf. They might be innately curious about what the generative AI had to say. All in all, if you are in need of topics to discuss on Valentine’s Day, you could end up using the generative AI topic per se and milk it for all that it is worth (perhaps replacing stony silence when you had nothing else in mind to discuss).
What might you do, either as someone who opts to use generative AI or someone who is told or finds out that the other beloved person used generative AI?
Grab yourself a glass of fine wine and mull over the enigma.
I have no way of knowing which of those above scenarios or situations is most applicable to you. One supposes that you will have to wrestle with what some would argue is an AI Ethics question. Should you be ethically bound to reveal that you have used generative AI, or are there circ*mstances whereby this does not rise to some ethical threshold that requires you to do so? For my ongoing coverage of the challenging arena of AI Ethics and also AI Law, see my analyses at the link here and the link here, just to name a few.
At this juncture, let’s get on with the show.
I will indicate a particular Valentine’s Day topic of interest and walk you through my use of ChatGPT. You will see first-hand what the world’s most popular generative AI has to say (note: an estimated 100 million weekly active users are reportedly making use of ChatGPT). Lessons will be sneaked into the discussion.
Here we go.
1. Knowing And Caring About What Valentine’s Day Is All About
Whenever you use generative AI to inquire or dialogue on a specific topic, you ought to first make sure that the generative AI app has already encountered the topic during the initial data training of the AI.
Allow me to explain.
Most generative AI is initially data-trained by scanning a vast array of essays, narratives, blogs, and all sorts of online text found on the Internet. The algorithm and data structures used for generative AI will attempt to pattern match onto the scanned wordings. As you might expect, the topics are wide-ranging and principally surface-level. By this, I mean that you should not assume that generic generative AI will have been data trained on in-depth domains such as medicine, legal and the law, etc.
Ergo, do not carelessly and mindlessly start using generative AI to aid you in any deeply steeped topics.
The AI makers warn you about this. They warn you by stating in their online licensing agreement that you have automatically agreed to properly use the AI when you sign up to use the app. I would guess that few people realize they have actively and unknowingly agreed to the legal terms and conditions. Regrettably, this means they people are often blindly using generative AI without knowing what is allowed and what is prohibited, plus, they don’t know what the AI can do versus what it is not data trained in. See my detailed coverage of those aspects at the link here.
I will begin my exploration of Valentine’s Day by asking what ChatGPT seems to be able to indicate about the famous occasion.
- My entered prompt: “Tell me about Valentine's Day and anything important that I need to know. Be brief.”
- ChatGPT generated response: “Valentine's Day, celebrated on February 14th, is a day dedicated to love and romance. It's a time when people exchange gifts, cards, and affectionate gestures with their significant others, friends, and family members. Roses and chocolates are popular gifts, and romantic dinners are common. It's essential to remember that Valentine's Day isn't just about romantic love; it's also a day to celebrate all forms of love and appreciation in your life.”
End of ChatGPT response
First, inspect the ChatGPT response and gauge whether it seems to make sense to you. I would suggest that the response here about Valentine’s Day is apt and correct. I am looking to see if anything odd is stated or otherwise gets my Spiderman Spidey sense tingling.
Here’s why you need to do this.
You might have heard that generative AI can produce something known as AI hallucinations. I must tell you that I disfavor the catchphrase as it implies that today’s AI is sentient and experiences hallucinations as humans do. That’s wrong. This is an unfortunate anthropomorphizing of AI. If you are further interested in what it is and how to deal with it, see my analysis at the link here.
Anyway, these so-called AI hallucinations entail the generative AI seemingly making up stuff that is fictitious and not based on factual aspects. It can readily happen. You need to always be on your guard that anything generated by the AI could have falsehoods, errors, made-up stuff, and the like. Closely inspect everything emitted. In this instance, suppose the depiction of Valentine’s Day had said that people dress up in scary costumes on this day (a false crossover about Halloween), or that Valentine’s Day occurs on March 30 of each year (an error, maybe made up).
I say over and again that generative AI is like a box of chocolates, namely that you don’t know for sure what you are going to get.
The second aspect that I’d like you to observe is the prompt that I used. When coming up with your prompts, you want to compose them sensibly. Be direct. If you are obtuse, the AI might not be able to sufficiently parse your prompt. There is an entirely new field of specialty known as Prompt Engineering that encompasses the best ways to formulate prompts, see my series at the link here and the link here.
Your prompt wording will make or break the value of using generative AI. Lousy prompts tend to get lousy results. Good prompts tend to get good prompts. When someone carps to me that generative AI wasn’t providing helpful responses, I usually discover that they are clueless about prompting. I lay that on their shoulders and not on the back of the generative AI.
Okay, we are ready to move on.
2. Suggestions For Valentine's Day Gifts
I am eager to find out what kind of Valentine’s Day gifts I should buy or prepare. There are only a few days left before the magical day. Maybe I’ve waited too long.
Let’s see what ChatGPT indicates.
- My entered prompt: “It is just about five days before Valentine's Day. Time is rushing along. I neglected to get anything for a loved one. What kind of a gift do you suggest that I buy or make? Give me some practical suggestions. Be brief but don't shortcut your answer.”
- ChatGPT generated response:
“If you're short on time, consider thoughtful gestures or experiences that show your love and appreciation. Some quick yet meaningful gift ideas include:”
“Handwritten Love Letter: Express your feelings in a heartfelt letter. It's personal and timeless.”
“Digital Photo Album: Compile memorable photos together and create a digital album or slideshow.”
“Homemade Treats: Bake cookies, make chocolates, or create a DIY gift basket with their favorite snacks.”
“Subscription Service: Sign them up for a subscription box related to their interests, such as gourmet food, books, or skincare.”
“Personalized Playlist: Curate a playlist of songs that hold special meaning for both of you.”
“Remember, the sentiment behind the gift matters most.”
End of ChatGPT response
Take a close look at the generated response.
You will notice that ChatGPT leaned in the direction of suggesting that I make something. That’s likely due to my prompt having emphasized the tightness of time. Do you see how your prompt can demonstrably impact the potential reply? I didn’t intend to have ChatGPT overlook my buying something. Nonetheless, that’s how my prompt was parsed.
What should I do now?
I could presumably use that stated response and try to do nothing further with ChatGPT. That would be a bit of foolishness on my part. I’d like to tell you why.
One really important part of using generative AI is that you need to keep in mind to undertake a dialogue or conversation with the AI. I say this because a lot of people often do a one-and-done approach. They would have stopped this session of using ChatGPT and walked away thinking that the only chance they have now is to make something.
I have noted in my column the truly valued ways that you use generative AI entails engaging in a dialogue with the AI. Here are my tips about doing so at the link here, thus a heads-up on avoiding mistakes that newbies experience when using generative AI at the get-go.
I will instruct ChatGPT that I still have time to buy something and see where this conversation might lead.
- My entered prompt: “I appreciate those gift ideas, thanks, but I believe that I do still have enough time to go over to the mall and buy something for my loved one. Thus, in addition to the gift ideas you've mentioned, what else might I get at a store? I have a limited budget so I need to be mindful of how much I can spend. I don't want to look like a cheapskate.”
- ChatGPT generated response:
“Certainly! Here are some affordable yet thoughtful gifts you can find at a mall.”
“Personalized Mug: Look for a store that offers custom printing services. You can have a mug printed with a special message, a favorite quote, or a memorable photo.”
“Scented Candle: Choose a candle with a soothing scent that your loved one enjoys. It's a simple yet thoughtful gift that can create a cozy atmosphere.”
“Jewelry: Look for simple and elegant pieces like a necklace with a meaningful pendant or a bracelet with a personalized charm.”
“Photo Frame: Select a stylish photo frame and include a special picture of the two of you together. It's a sentimental gift that they can display in their home or office.”
“Book or Journal: If your loved one enjoys reading or writing, pick out a book by their favorite author or a beautiful journal for them to jot down their thoughts.”
“Scarf or Hat: Find a fashionable scarf or hat that matches their style. It's a practical yet stylish gift, especially during the colder months.”
“Remember, it's the thought and effort you put into the gift that truly matters, not the price tag. Choose something that reflects your loved one's interests and personality, and they're sure to appreciate it.”
End of ChatGPT response
As always, make sure to take a close look at the responses and try to discern whether there is anything oddball or untoward in the reply. I think this response seems sensible to me. Those are items that are likely within my limited budget, depending on what I can find at the mall.
You might be tempted to think that these responses are not particularly better than if you went onto the Internet and did a search for tips and suggestions about Valentine’s Day gift getting. I can see why you might think this.
The deal is that I could keep going in this ChatGPT conversation and become increasingly specific. For example, I might at this juncture home in on getting a scarf. I could provide details about the person that I want to get the scarf for. In turn, the generative AI will get more detailed on the suggestions about the gift.
It is harder to do the same conversational action on a traditional search engine (many of them are now being armed with generative AI, blending the two together). My gist here is that, unlike a conventional Internet search, the beauty of using generative AI is that you can carry on a dialogue and keep moving around, getting deeper or shifting directions, which is not something you can conventionally do in a typical search engine.
I do want to mention an allied matter to keep you on your toes.
Please realize that most of the generative AI apps do not guarantee you privacy or confidentiality, see my coverage at the link here. You need to assume that anything you enter as a prompt might be seen by the AI developers or used in other ways. Be cautious in what you say in your prompts, such as avoiding entering any personally identifiable information. Also, find out what stated rules or policies the generative AI that are you using has about such matters.
3. Celebrating Valentine’s Day
Now that I have some ideas about a gift, I need to be thinking about the magical day itself.
Let’s ask ChatGPT.
- My entered prompt: “I want to do something special on Valentine's Day. One problem is that we both work full-time and can't get together during the daytime. For the nighttime, we both have long commutes so we can't meet someplace until probably 7 p.m. or later. On top of that, I've heard that most of the restaurants in town are completely booked that evening. I should have booked something earlier, but I didn't. I am at my wit's end. What do you suggest we do as a means of celebrating Valentine's Day when we are faced with the aforementioned constraints?”
- ChatGPT generated response:
“Given your time constraints and the likelihood of restaurants being fully booked, you can still have a memorable and intimate Valentine's Day celebration with a creative and thoughtful approach. Here are some suggestions.”
“Cook a Special Dinner Together: Instead of going out, plan a romantic dinner at home. Choose a simple yet delicious recipe that you both can prepare together. Set the mood with candles, soft music, and a nicely set table.”
“Picnic Indoors: Create an indoor picnic by spreading out a blanket on the living room floor. Prepare a selection of finger foods, cheese, crackers, fruits, and maybe a bottle of wine. Enjoy a cozy and intimate meal together.”
“Homemade Spa Night: Pamper each other with a DIY spa night at home. Set up a relaxing atmosphere with candles, essential oils, and soothing music. Give each other massages, do facials, and enjoy a relaxing bath together.”
“Movie Marathon: Choose a few of your favorite romantic movies or a series to binge-watch together. Make some popcorn, cuddle up on the couch, and enjoy a cozy night in.”
“Exchange Personalized Gifts: Take the time to exchange thoughtful and personalized gifts that show your love and appreciation for each other. It could be something handmade, a scrapbook of memories, or a special item that holds significance for both of you.”
“Plan a Future Date: Use Valentine's Day as an opportunity to plan a future date together. Research and make reservations for a special outing or experience that you can enjoy when you have more time.”
“Remember, it's the thought and effort you put into spending quality time together that truly makes Valentine's Day special. Enjoy each other's company and cherish the moments you share.”
End of ChatGPT response
In this case, I am using generative AI to get personalized advice. You might say this is akin to writing a letter to Dear Abby (well, that’s maybe showing my age).
I think the generated advice makes sense. As mentioned repeatedly, first do a sensibility check on what the response has to say. After mulling it over, I have decided that I am going to make use of the picnic indoors. That will allow us to both come home and not have to battle to park and find an available eatery. I am going to get the indoor picnic items suggested by ChatGPT.
I also like the idea of doing a movie marathon. There are several movies we’ve wanted to watch but just hadn’t taken the time to do so. Maybe now is the time to watch them.
Wait for a second, we both will need to work the next day, so a movie marathon is pretty much out of the question. I can though plan on us watching a single movie. We have time for that.
I mention this because I want to emphasize something here that is just as important as the other tips, that I’ve noted for you.
Here it is, get ready.
You do not need to abide by whatever generative AI tells you.
For example, I nixed the idea of a movie marathon but meanwhile morphed it into the notion of watching a single movie. That’s perfectly fine. I can do what I want with whatever the generative AI perchance happens to say. I’m sure you are probably thinking, of course, you can do whatever you want. Believe it or not, lamentedly, some people begin to fall into a kind of trance and believe whatever generative AI tells them. I have written extensively about the mental health concerns associated with using generative AI.
Generative AI can be a really big help to mental health and provide all kinds of useful guidance and advice. In the same breath, please know that generative AI can undercut mental health and we need to be watchful of where things are heading. For my explorations on generative AI and mental health, see the link here and the link here, among many of my assessments.
4. Boosting Romantic Prowess
Things are starting to take shape for my Valentine’s Day plans.
One issue that I know might arise is that I am not much of a romantic. Perhaps having spent so much time around computers and AI has led me down a drier path. What am I to do?
Time to ask ChatGPT.
- My entered prompt: “I like your suggestions. I am going to have an indoor picnic. I will also kind of do a movie marathon but we only will have time to probably watch just one movie since we need to go to work the next day. I am admittedly not much of a romantic person, but I want all of this to have a semblance of romance. What do you think I can do given the choices I've made about the evening to make it romantic in tone?”
- ChatGPT generated response:
“That sounds like a wonderful plan! To add a touch of romance to your indoor picnic and movie night, here are some suggestions.”
“Ambiance: Set the mood with soft lighting, such as candles or string lights. Dim the main lights to create a cozy atmosphere. You can also use scented candles or diffusers with romantic fragrances like lavender or vanilla.”
“Decor: Spruce up your indoor picnic area with a nice tablecloth or blanket. Add some fresh flowers or rose petals for a romantic touch. Consider using decorative pillows or cushions to make the seating area more comfortable and inviting.”
“Music: Play some soft, romantic music in the background to enhance the mood. Choose songs that are meaningful to both of you or opt for instrumental music for a more relaxed ambiance.”
“Personal Touches: Incorporate personal touches to make the evening special. Write love notes or romantic messages to each other and hide them around the picnic area for a sweet surprise. You could also include small keepsakes or mementos that hold significance for your relationship.”
“Food and Drinks: Select food and drinks that you both enjoy and that feel indulgent. Consider splurging on a bottle of champagne or sparkling cider to toast to your love. Don't forget to include some decadent desserts to satisfy your sweet tooth.”
“Movie Selection: Choose a romantic movie that you both love or one that holds special meaning for your relationship. Snuggle up together under a blanket while watching the film and share moments of laughter and tenderness.”
“By incorporating these elements into your indoor picnic and movie night, you'll create a romantic and memorable evening that celebrates your love and connection. Enjoy the time together and cherish the moments you share!”
End of ChatGPT response
I appreciate those answers.
You might be exhorting that I should have thought of those on my own. Sure, I get that. But, hey, my mind is focused on software engineering and not on setting up a romantic atmosphere. Let me be.
I will light some candles that we have in the house. I’ll get some fresh flowers at the store on the way home that evening. I’ve got a playlist of songs that we both relish. I know exactly what kind of food would go with the occasion. For the movie, I’ll provide a list of three that are rom-coms and see which one is considered most desirable for watching that evening.
If I needed it, I could ask ChatGPT to produce a checklist of what I have to assemble. I could also tell ChatGPT to produce a timeline for Valentine’s Day that would show what things I have to get done before the evening rolls around.
I won’t show you that here. Just be aware that those are the kinds of things that you can have generative AI produce.
Many people seem to falsely assume that generative AI will only produce essays or conventional text. They don’t grasp that you can use generative AI to produce checklists, timelines, schedules, and even charts and other formatted facets (I refer to this as Target-Your-Response or TPR prompting, see my description at the link here). That’s why learning about prompts and prompt engineering can be advantageous.
5. Being Conversational During Valentine’s Day
Get ready to roll your eyes.
When I have free time, which I don’t, but when I do, I still tend to think about and talk about AI and computing. Perhaps it is in my blood. Somehow, I just always stay in that mode. Can’t break out of it.
For the indoor picnic, along with the flowers and the rest of the romantic ambiance, I could spoil the whole kit and kaboodle by nonstop conversing about the latest in AI. What else can I bring up? I could talk about what’s in the news overall but given that the news these days is rather heavy, I don’t think that’s the heartwarming dialogue that befits the setting.
What in the world will I discuss?
Let’s ask ChatGPT.
- My entered prompt: “I am going to use candles to set the ambiance. I will make sure to get some fresh-cut flowers beforehand. I have a music playlist that will work perfectly. All in all, I think the approach outlined is good. But, I am not sure what we might talk about on that evening. I tend to be focused on my work. What ideas or suggestions do you have about how I could be appropriately conversational for that magical evening?”
- ChatGPT generated response:
“It's great that you're thinking ahead about conversation topics to make the evening enjoyable and engaging. Here are some suggestions to keep the conversation flowing.”
“Reflect on Memories: Take the opportunity to reminisce about happy memories and experiences you've shared together. Share stories about how you met, your first date, or memorable adventures you've had.”
“Ask Open-Ended Questions: Encourage meaningful conversation by asking open-ended questions that invite your partner to share their thoughts and feelings. For example, "What's something you've always wanted to do but haven't had the chance to yet?" or "What's your favorite thing about our relationship?"
“Discuss Dreams and Goals: Talk about your hopes, dreams, and aspirations for the future. Share your personal goals and ambitions, and ask your partner about theirs. This can lead to insightful discussions and strengthen your connection.”
“Play a Game or Trivia: Incorporate fun and interactive activities into your evening, such as playing a board game or trivia game together. It's a lighthearted way to spend time together and spark conversation.”
“Express Gratitude: Take the time to express gratitude for each other and the relationship you share. Share what you appreciate most about your partner and how they enrich your life.”
“Talk About Shared Interests: Discuss topics that you both enjoy and are passionate about. Whether it's travel, hobbies, or favorite books and movies, sharing common interests can lead to engaging conversations.”
“Remember to be attentive and actively listen to your partner's responses. The goal is to connect on a deeper level and enjoy each other's company during this special evening. Enjoy the conversation and the time spent together!”
End of ChatGPT response
I will once again pick and choose from the options presented.
In terms of memories, I think that I can easily put together digital photos of us and be ready to go through those. Plentiful conversation is to be had. We can reminisce about those wonderful moments. I’m sure doing so will also inspire us to talk about some upcoming vacation trips we’ve been contemplating.
I also like the suggestion of sharing gratitude. In our hectic lives, we often forget to mention appreciation for each other. It is an easy trap to fall into. I suppose that Valentine’s Day is an ideal time to be reminded of the importance of shared gratitude. Nice touch!
One thing you might be wondering about is that I keep telling the generative AI what I am going to do and how I cherrypicked from the suggestions that were generated.
Why would I do this?
The answer is simple and straightforward.
If you want to keep a conversation going, make sure to let generative AI know what you are thinking about. I say this because you have to realize that generative AI is not a mind reader (well, not yet, though efforts of connecting AI to human brains are underway and known as Brain Computer Interface or BCI, see the link here).
By failing to mention in your prompts what you are deciding, you will leave the generative AI in the dust and no longer be aligned with what you are trying to accomplish. The subsequent responses will seem increasingly less instructive to you. A gap is forming between what you’ve told the AI versus what is bouncing around in your noggin. It is useful therefore to provide quick updates in your prompts to ensure that the AI computationally keeps up with your line of thought.
6. Don’t Ruin The Vaunted Valentine’s Day
So far, I’ve been identifying a handy plan of action for Valentine’s Day.
Feeling good, pretty good.
You might be familiar with an old line about battle plans. The line is that no plan fully survives the moment that the first shot is fired. The chances are that a plan will be waylaid or must adapt to the situation at hand as things unfurl.
Another way to look at this quandary is that I don’t want to be the one who ruins Valentine’s Day. Not me, by gosh. But, as they say, the problem is that I don’t know what I don’t know. I know what I know. And I know what I don’t know. The big gaping hole is what I don’t know which I don’t even know.
A tongue twister.
Perhaps generative AI can assist in my quest to avoid ruining things.
- My entered prompt: “I have now shared with you my overall plans for Valentine's Day. I have chosen various options that you suggested, thank you. My question now is what might ruin this plan? What should I be wary of or try to avoid?”
- ChatGPT generated response:
“It's great that you're considering potential pitfalls to ensure your Valentine's Day plans go smoothly. Here are some things to be wary of and try to avoid.”
“Overly High Expectations: While you've put effort into planning a special evening, try not to set unrealistic expectations. Remember that the most important thing is spending quality time together, rather than the perfection of every detail.”
“Communication Breakdown: Make sure you communicate effectively with your partner about your plans and expectations for the evening. Miscommunication or assumptions can lead to disappointment or frustration.”
“Time Management: Be mindful of time constraints, especially if you both have busy schedules. Plan your activities accordingly to ensure you have enough time to enjoy each other's company without feeling rushed.”
“Logistical Issues: Anticipate and address any potential logistical challenges, such as traffic delays, last-minute shopping, or technical difficulties with your movie setup. Have a backup plan in place to handle unexpected issues.”
“Neglecting Personal Touches: While planning is important, don't forget to add personal touches and spontaneity to the evening. Be present in the moment and focus on connecting with your partner rather than just going through the motions of your planned activities.”
“External Distractions: Minimize distractions during your time together by silencing phones, turning off notifications, and creating a peaceful environment. This will help you fully immerse yourselves in the experience and enjoy each other's company without interruptions.”
“By being aware of these potential pitfalls and taking proactive steps to address them, you can help ensure that your Valentine's Day plans are successful and memorable for both you and your partner.”
End of ChatGPT response
What do you think of the response?
For me, there is a gigantic point that I sheepishly would have likely overlooked. The point was made that there should be room for spontaneity. Got me on that one. I was going to put this on a detailed and strict timeline that would have us doing each thing on a minute-by-minute basis. An order of march, one might say. I already started mapping this in a spreadsheet and was thinking of using a project management tool to prepare a Gantt chart.
It would have been Valentine’s Day via stopwatch.
Maybe not the most romantic splendor that could be envisioned.
Spontaneity to the rescue, as noted by generative AI.
7. What If Valentine’s Day Goes Amiss
I promised you that there would be seven examples, and here’s the seventh one.
We are nearing the end of this joyful exploration of generative AI for use in preparing for Valentine’s Day. My lucky number seven led us to adventurously examine seven examples. Plenty more could be derived. You might want to try your own luck and use generative AI to aid in your Valentine’s Day planning too.
This seventh one might seem a downer, but I don’t intend it to be that.
Suppose that something goes amiss during Valentine’s Day. There, I said it. You might insist that I have now jinxed myself. I said the quiet part aloud. Had I not said anything about things going awry, the occasion would have been flawless. I doomed myself.
A counterargument is that just as ChatGPT noted, there is always a chance of the best-laid plans going topsy-turvy. I want to mentally prepare myself for the worst-case scenarios. If somehow and unimaginably the night turns out disastrous, how will I pick myself up and continue my day-to-day existence?
Okay, that was over the top, sorry. Didn’t want to make things seem that bad. I doubt very much that the evening will have any material divots. If something adverse happens, it will almost certainly be inconsequential in the big scheme of things.
Here’s what we can do.
I will ask generative AI to do a pretend. We will pretend that on Thursday, February 15, I logged into ChatGPT and said that things had some rough edges. I want to see in advance what generative AI can do to help me out of a slump.
- My entered prompt: “I want to do a pretense with you. Let's pretend that I come to you on the day after Valentine's Day. I am sad and upset because the evening did not go as planned. I am feeling low. What could you do to walk me through my low feelings? I want to carry on a pretend dialogue with you. You are to first ask me how I am feeling. I will tell you that I am feeling sad because of what happened on the evening of Valentine's Day. You then will interact with me, consoling me as we discuss how I am doing. Do you understand these instructions?”
- ChatGPT generated response: “Yes, I understand the instructions. Let's begin our dialogue.”
- My entered prompt: “It’s the day after Valentine’s Day and I’d like to chat with you about something.”
- ChatGPT next generated response: “How are you doing today?
- My entered prompt: “Not great. I'm feeling sad and upset about how Valentine's Day turned out.”
- ChatGPT generated response: “I'm sorry to hear that. Would you like to talk about what happened? Sometimes sharing our feelings can help lighten the burden.”
- My entered prompt: I wanted to make sure we had a perfect evening. Things seemed to go wrong at every turn. The candles were overpowering. The indoor picnic was a bit underwhelming. We had trouble deciding which movie to watch. Frankly, I think the main issue was that we both were just so tired from work. We usually wait until Saturday to do anything out of the ordinary, giving us time to relax and unwind. A Wednesday evening is not conducive to this since we still were faced with working on Thursday and Friday.”
- ChatGPT generated response: “It sounds like you had high hopes for the evening, and it's disappointing when things don't turn out as expected. But remember, it's okay for things to not always go according to plan. What's important is that you and your beloved were together and made an effort to celebrate your love, even if it wasn't perfect.”
- My entered prompt: “I just wanted it to be a very special occasion.”
- ChatGPT generated response: “I understand. It's completely normal to feel disappointed when our expectations aren't met. But perhaps there were still some moments of connection or laughter that took place throughout the evening that you can hold onto. And remember, there will be many more opportunities to create special moments together in the future.”
- My entered prompt: “That makes sense. I will keep that in mind today and probably by this weekend I will be feeling much better. Perhaps on Saturday, Valentine’s Day will be a fond memory and the freshness of things that went awry will be a distant memory.”
End of ChatGPT dialogue
I’d like to cover a few final points that are raised by the above dialogue and then we’ll move on to the final remarks of this discussion.
First, you might have observed that generative AI was almost acting in a counseling or coaching manner. I was asked questions and/or given advice or guidance. Do you think this constitutes providing mental health guidance or therapy? Some do, and they are concerned that this is being done without any human therapist being in the loop. Others claim that this is fine, and available 24x7, at a relatively low cost, so this allows everyone to have access to mental health therapy.
For my probing discussions on the contentious topic, see the link here and the link here.
Second, generative AI was able to engage in a pretense, often labeled as making use of a persona. One of the most overlooked prompting strategies is that you can readily get generative AI to take on a persona. Suppose a student in school wants to see what it might have been like to interact with President Abraham Lincoln. You can have generative AI simulate that persona. It won’t be perfect, and you’ll need to realize it is just a pretense, but in any case, the capability can be quite handy.
For the keystones about prompting of personas in generative AI, see the link here and the link here.
Third, the wording emitted by generative AI often uses the word “I” as though the AI is a sentient being. We associate the word “I” with usage by humans. When it is shown by AI, we tend to be lulled into thinking that the AI is sentient.
I refer to this as the AI makers doing anthropomorphizing by design.
You might be shocked to know that there is no requirement that the AI makes use of the word “I” in this manner. The AI makers could easily stop this. They have tilted the AI in that direction, purposefully so. The claimed logic is that the AI seems friendlier, and people will more readily use the AI. A retort is that it is a sneaky trick to convince people that AI is more than it seems. The trickery is done to gain views and make sure that people keep coming back to use generative AI.
To unpack a quick example, one of the above lines says this: “I understand. It's completely normal to feel disappointed when our expectations aren't met.” The word “I” is not required. The sentence could have started with “This is understandable.” See how dramatically that changes the intonation and emotional draw. No longer is there an “I” reference that was otherwise not needed.
Please do not allow yourself to be taken in by the aura of sentience. We do not have sentient AI. Despite any headlines that proclaim that we do, let it be known that we don’t.
Conclusion
Congratulations, you made it through a planning process for Valentine’s Day and simultaneously learned what I ardently believe are crucial insights for you about the nature of generative AI.
It was in fact a twofer.
I sincerely hope you will have a wonderful Valentine’s Day.
Henry David Thoreau, the great American essayist, said this about love: "There is no remedy for love but to love more." It’s a momentous comment and one that maybe we can dream to be true.
Roses are red, violets are blue, and may generative AI be useful to you.
