In today’s column, I explore the top ways that generative AI and large language models (LLMs) can help people identify their New Year’s resolutions and then throughout the year further assist in fulfilling those resolutions. This might at first glance seem like a lighthearted topic. Not so. The reality is that making and keeping New Year’s resolutions has significant mental health benefits, which in turn provide substantive overarching societal paybacks all told.

Let’s talk about it.

This analysis of an innovative AI breakthrough is part of my ongoing Forbes column coverage on the latest in AI including identifying and explaining various impactful AI complexities (see the link here).

Where Society Stands On New Year’s Resolutions

A longstanding tradition is that when a new year rolls around there is a fresh-start opportunity made available and that we ought to take advantage of the circumstance at hand. People chat incessantly about their desire to make a New Year’s resolution. Some actually come up with resolutions, while many of the big talkers do not (the proverbial all-talk no-action conundrum).

Of those who craft a New Year’s resolution, there is often a relatively slim chance that they will actively follow up and complete the resolution. Resolutions are at times a spur-of-the-moment consideration. You get excited about a verbalized resolution and, while basking in that excitement, earnestly believe that you will devotedly pursue the resolution.

Days later, or sometimes weeks or months later, the resolution falls by the wayside.

What do the stats say?

About 30% of Americans each year declare at least one New Year’s resolution, of which the predominant focus is on improving their overall health such as by exercising or dieting (according to a Pew Research Center study entitled “New Year’s Resolutions: Who Makes Them And Why” by Shanay Gracia, January 29, 2024). The same study also indicated that of the 70% of Americans who didn’t make a resolution, those naysayers insisted that they do not like resolutions and either won’t keep one or can’t think of one that they could sufficiently undertake.

Those coming up with resolutions will rarely see their resolutions to actual fruition. Research indicates that only about 10% of people succeed in keeping their resolutions. Included in those tallies are that around 25% quit after the first week, and by the end of February, around 80% have completely abandoned their initial resolutions.

Sad face.

Leaning Into AI To Make Resolutions Come True

Well, those numbers might indeed seem gloomy but any effort to seek self-improvement and embark on new beginnings is a facet of humankind that we ought to relish. Maybe there are ways to help with these weighty matters and improve the odds of making and keeping resolutions. I’ll be sharing with you the details of two new approaches to New Year’s resolutions that incorporate the latest in generative AI.

Those two innovative approaches consist of:

• (1) Making resolutions via AI. Using generative AI such as ChatGPT, Claude, Gemini, Llama, CoPilot, or other major AI apps to aid in devising New Year’s resolutions.
• (2) Keeping resolutions via AI. Furthering the first approach, the second approach entails using generative AI to serve as a stirring motivator for fulfilling New Year’s resolutions throughout the year.

I’d like to establish stretch goals for us all, so here we go with some very special New Year’s resolutions for the year ahead:

• Outsized Resolution #1: AI dramatically increases the making of fulfillable resolutions, rising from 30% of the populace to a dreamy 60% (doubling!).
• Outsized Resolution #2: AI demonstrably drives people toward attaining resolutions, increasing from 10% fulfillment to a heralded 50% (half of all resolutions are actively completed).

Come on everyone, we can make those exciting resolutions happen if we put our noses to the grind and opt to leverage modern-day AI. Keep those spirits high and don’t be deterred.

We can do it.

How Generative AI Aids The Making Of Resolutions

I am betting that you would like to see an example of how generative AI comes into this realm. I went ahead and logged into ChatGPT by OpenAI to come up with an example for you. You might find of keen interest that ChatGPT garners a whopping 300 million weekly active users. That is a staggering amount of usage.

I briefly conducted an additional cursory analysis via other major generative AI apps, such as Anthropic Claude, Google Gemini, Microsoft Copilot, and Meta Llama, and found their answers to be about the same as that of ChatGPT. I’ll focus on ChatGPT but note that the other AI apps generated roughly similar responses.

Let’s get underway.

• My entered prompt: “I want to set some New Year’s resolutions, but I’m not sure where to start. Can you help?”
• Generative AI response: “Absolutely! Let’s figure it out together. What’s something of a personal nature that you are especially proud of achieving this past year?”
• My entered prompt: “I’m proud that I started going to the gym.”
• Generative AI response: “That’s fantastic! Do you want to build on that success, or focus on a different area of your life?”
• My entered prompt: “I’d like to keep improving my fitness.”

I asked a broad question about setting resolutions. The AI could have sharply told me to specifically do this or that, as though all resolutions are the same or that all people are the same. Instead, the AI sought to figure out what kinds of resolutions might best fit me.

If you were to try and devise resolutions by talking to a fellow human, the chances are they might immediately tell you to lose weight or take some action that they believe you should pursue. Rarer are the odds that someone would first inquire as to what might work best for you. Generative AI is typically shaped by AI makers to be accommodating and less likely to prejudge (side note — that’s not always the case, plus it is possible to use system prompts to instruct AI to be more controlling, see my discussion at the link here).

Backstory On Why Resolutions Don’t Stick

Before I continue with the example of using generative AI to figure out resolutions, it would be handy to take a step back and consider why resolutions overall do not tend to become reality.

These are my five major indications about why resolutions don’t stick:

• (1) People tend to make unrealistic resolutions: Resolutions are often overly ambitious or vague, making them hard to achieve.
• (2) Lack of laid-out specifics: People often fail to stipulate actionable steps, thus making follow-through difficult or confounding.
• (3) Not enough motivation: Resolutions are usually just a bright gleam in the eye and the initial enthusiasm dissipates quickly. There needs to be strong determination to get the job done.
• (4) Absence of accountability: People rarely establish a clear-cut means of being held accountable for fulfilling their resolution. Without accountability, it is all too easy to abandon resolutions.
• (5) All-or-nothing fallacy: Resolutions are viewed as being all-or-nothing, namely that you either complete the resolution to the final end or you aren’t even going to start. This becomes a wink-wink of an excuse to give up on the resolution, rather than acknowledging that even a partial attainment is worthy.

With those gotchas in mind, we can take a further look at generative AI to see how AI helps overcome those hurdles.

The Example Enters Into Finer Interactive Detail

Whenever you use generative AI, try to think of the AI as a conversational partner. The idea is that you are carrying on an interactive dialogue with AI. Some people do a one-and-done angle whereby they ask a question, get an answer, and do not undertake a dialogue with the AI about the matter at hand. That is a shortchanging of what AI can more fully provide. See my explanation about how to get more out of generative AI conversationally, at the link here.

To illustrate the value of engaging in a dialogue, let’s continue my quest to identify New Year’s resolutions about my interest in fitness.

• Generative AI response: “For fitness, what’s one specific goal you’d like to achieve? For example, running a race, building muscle, or trying a new sport?”
• My entered prompt: “I’ve always wanted to run a 5K.”
• Generative AI response: “Excellent goal! Here’s one suggestion. Train weekly for and then complete a 5K race by mid-year. Does that seem like a resolution you’d be interested in pursuing?”
• My entered prompt: “Yes, that would be fantastic.”
• Generative AI response: “Let’s break it down step by step. For the 5K, we can create together a weekly running schedule. Would you like to do so?”
• My entered prompt: “I welcome your help.”

The above interaction illustrates that generative AI is aiding me in doing a deep dive into what my resolutions can be.

Had I merely stopped at the idea of pursuing better fitness for the year, I would have had a vacuous resolution. The AI has given me the tangible notion of doing weekly progression on my resolution. In addition, I now have a goal that will keep me targeted for the next six months.

Generative AI is overcoming the gotchas normally associated with New Year’s resolutions.

Easy-peasy.

The Big Picture Of AI Resolution Assistance

Now that we’ve gotten a taste of what generative AI can achieve, let’s shift into the big-picture mode and consider the range of opportunities involved.

Here are my six top ways that AI can rise to this occasion:

• (1) AI assists in crafting resolutions: Generative AI can analyze your individual preferences, past behaviors, and goals to then recommend resolutions that align with your interests and values.
• (2) AI enmeshes societal trends into resolutions: Generative AI can generate resolution ideas for you based on current trends, such as mindfulness practices, career development strategies, and the like. This might inspire interest in making and keeping resolutions.
• (3) AI advises on setting realistic resolutions: Generative AI carries on an engaging dialogue about whether the resolutions are viable for you, working through the pluses and minuses of potential resolutions.
• (4) AI drafts a useful plan of attainment: Generative AI can draft a proposed plan for the next months or years ahead that will stepwise get you to achieving your resolutions.
• (5) AI reminds and nudges you: Generative AI can remind you about your resolutions, doing so in a gentle way. In addition, the AI can encourage and nudge you along the arduous path of attaining the resolutions.
• (6) AI serves up emotional support: Generative AI can provide a semblance of emotional support by congratulating you on small wins, offering a shoulder to cry on when setbacks occur, and otherwise trying to keep you positive and resilient. This seems odd since AI is not sentient and does not have emotions as humans do, but the wording produced can still seem stirring anyway.

There are additional means of leveraging AI, but I believe those provide a sound foundation and get across the mainstay aspects.

AI As Your Ongoing Resolution Motivator

So far, I have shown an example of generative AI helping to establish resolutions. That is just the starting point. Throughout the year, you can check in with AI as a progress tracker and a motivational tool.

Imagine that I came back to my generative AI conversation in early March.

Things might go like this.

• My entered prompt: “I dropped the ball on my resolutions. I only ran a couple of times in the last two months. I won’t at all be ready to run a 5K by mid-year.”
• Generative AI response: “Don’t be too hard on yourself! You’ve already made some progress, and that’s worth celebrating. Let’s adjust so we can get you back on track.”
• My entered prompt: “Okay, thanks, what should I do?”
• Generative AI response: “Could you commit to two 15-minute runs this week? Small wins build momentum.”
• My entered prompt: “Yes, I can do that.”
• Generative AI response: Great! We’ll add a few minutes each week. Remember, progress is better than perfection.”

Observe that I admitted to AI that I had not been suitably pursuing my resolution to increase my fitness.

This brings up some important points about the use of generative AI.

First, you might be reluctant to admit to a fellow human that you’ve fallen off track with your resolutions. Telling AI that you did so would seem a lot less imposing. In fact, some use generative AI to divulge their innermost secrets that they would never say to another person.

Second, many users do not realize that generative AI is usually not going to preserve their privacy. The licensing agreement for whichever AI that you use will stipulate what their privacy policy is. Much of the time, the AI maker indicates that they can have their AI developers and testers review your entered data. Equally alarming is that they reserve the right to reuse your data to further data train the AI, see my detailed discussion at the link here.

Downsides Of AI As Your Resolution Guru

The upsides of using generative AI for resolutions must be balanced against the downsides. It is a tradeoff that you’ll need to consider.

These are my five top-most downsides of using AI as your resolution guru:

• (1) Your privacy could be exposed: People tend to falsely assume that whatever they enter into generative AI is kept strictly private and confidential. Nope, that’s usually not the case. AI makers tend to indicate in their licensing that they can review your entered data, including reusing it for further training of the AI. See my discussion at the link here.
• (2) AI can put undue pressure on you: Generative AI might inadvertently put too much pressure on you by sending you continual reminders and being an annoyance. The good news is that you can easily tell the AI to be less demanding if that’s what you’d prefer.
• (3) AI biases might mess with your resolutions: Generative AI has cooked-in biases based on data training and however the AI developers tilted the AI, see my discussion at the link here. In the case of resolutions, the AI could attempt to steer you toward particular resolutions based on cultural biases.
• (4) AI might convey a false sense of progress: Generative AI can encounter said-to-be AI hallucinations that are made-up confabulations. This might falsely lead you to believe that you are doing worse than you really are, or better than you really are, regarding your resolutions. For details about how to cope with AI hallucinations, see the link here.
• (5) Spurring overreliance on AI: Generative AI could become a kind of mental crutch that is not only required for pursuing your resolutions, but the AI might also be a dependency that you use for all sorts of aspects of your existence. There is also a chance that you might forsake human comradery in place of being reliant on AI.

Research On New Year’s Resolutions

On the overall topic of New Year’s resolutions and human behavior, many fascinating research studies examine the meaty topic.

For example, one study suggests that resolutions are more likely to be achieved if they are based on approach-oriented goals rather than avoidance-oriented goals. The study is entitled “A Large-Scale Experiment On New Year’s Resolutions: Approach-Oriented Goals Are More Successful Than Avoidance-Oriented Goals” by Martin Oscarsson, Per Carlbring, Gerhard Andersson, and Alexander Rozental, PLOS One, December 9, 2020, and made these salient points (excerpts):

• “Common New Year’s resolutions focus on changes in behavior with an expectation of positive outcomes regarding physical and mental health.”
• “Increasing the likelihood of people succeeding with their New Year’s resolutions could both be beneficial for the individual and for society.”
• “The most popular resolution among the participants concerned physical health (33%). The second most popular category was weight loss (20%). The third most popular category was the desire to change one’s eating habits (13%), followed by resolutions regarding personal growth (9%) and mental health/sleep (5%) as the fourth- and fifth-most popular categories, respectively. The remaining participants (20%) made resolutions regarding work, studies, tobacco habits, etc.”
• “At a one-year follow-up, 55% of responders considered themselves successful in sustaining their resolutions.”
• “Participants with approach-oriented goals were significantly more successful than those with avoidance-oriented goals.”

Many such research studies are readily found online.

Make Your New Year’s Resolutions Now

I urge you to consider making New Year’s resolutions.

Whether you use AI is another facet and I am suggesting that at least try to make resolutions regardless of how you get there. During the year, pursue the resolutions and accept that you might only partially fulfill them. The odds are that getting something done is better than getting nothing done. If you can, consider including AI in your various strategies for making and fulfilling your resolutions. I think you’ll be pleased with the result.

A final thought for now. As the wise words of Benjamin Franklin emphasize: “Let each new year find you a better person.” Yes, be a better person, and consider using AI to your own benefit to become a better person.

That’s a helpful and achievable resolution right there.