Your PERSONAL AI Trainer ... HAS TO be YOU.
TL;DR: AI ain’t gonna cut it until people understand that AI is not PERSONAL enough.
Let’s say, just for the sake of argument, that we wanted to develop a Bruno.Solutions open source version of a personal *AI trainer, which that might be somewhat similar to Ditto.Fit
What would be the best way to go about this task? Well … MAYBE … we should just sign up for Ditto.Fit’s beta waitlist and see what they have to offer. Then we can decide if we want to build something similar. Or better. OR maybe we don’t want to build anything at all. Just yet.
The REAL problem is that we would need to find a way to overcome the oxymoronity of the expression personal AI … it’s not that there aren’t competitors in the space, but it’s just that the whole idea of a personal AI breaks down because that expression is a bit of a contradiction in terms … unless we find a way to make the plain Jane Fonda VCR-tape-style of AI fitness apps somehow more personal … HOW is the AI thingey going to drag my lazy ass up-and-at-it for my four-minute Tabata HIIT snacks that I should be doing 15X/day to 30X/day.
How does AI manage to get PERSONAL?
Good question!
Let’s look a the direct competitors Ditto.Fit:
Fitbod: AI-powered personal trainer focusing on strength training and weight lifting … the weights ain’t gonna lift themselves, maybe the app does it?
Centr: Fake-personalized workout plans by Chris Hemsworth’s team of celebrity trainers like LeBron James and Tony Horton … is there a way that the AI app can better signal that I am just not close to being stupid enough to fall for celebrity bullshit, especially about what skunkpiss celebrities want to spray into the world, ANY MORE than if I just don’t sign up? ONE thing, I do notice about this app, like ALL other apps or gym memberships or other fit-celeb products out there is how ESPECIALLY IMPERSONAL and FAKE celebrity bullshit is … BUT even though the staged photos look earnest, it’s clear as hell from looking at the terms that it will be 10X-level-impossible to cancel the damned parasitic shit once the outfit has your credit card.
Freeletics: AI-powered coaching for bodyweight workouts and HIIT … maybe I didn’t accept enough cookies or brownies or whatever the people who built this were thinking, but how this app is different than just googling body weight exercises just does not BEGIN to make sense to me and, somehow, I always understood the importance of basic bodyweight workouts in folks in prison, even solitary confinement in really nasty prisons.
Strava means “strive” in Swedish: Focuses on running and cycling, using AI to analyze performance and suggest workouts … maybe I don’t understand Swedish, but … so what?
Pacer for Teams … not the popular and useful US Courts PACER or the AMC Pacer that got Garth and Wayne out of the basement in Wayne’s World: This is an AI-powered running coach that provides personalized training plans and audio coaching … YAWN … but there’s more. There’s an example of a TEAM-based fitness app which matters because there actually are some organization leaders and HR staff who are really starting to get that far, far, far too many employees feel disconnected & overwhelmed & depressed & burdened by stress & likely to eat/avoid exercise. The TEAM-based or buddy-based approach works because it IS *personal. The buddy-based fitness practice is a great idea that actually has been proven to work, if only for making buddies into better buddies and more likely to actually call out the bullshit, rather than just accepting whatever someone says. It’s not exactly AI, but maybe AI can help be the basis of what makes teams be better teams and buddies be better buddies.
Similar ventures with different approaches:
Peloton: Offers live and on-demand fitness classes with virtual instructors … it’s a $2,000+ fashion statement that more than likely to to be used as a clothes drying rack in 6 months or 6 weeks … but it’s also kind of a cult of people who get suckered into throwing a couple grand at meaningless fashion statements, so the Apple cult does have some competition cult-wise.
Big Mirror … or, if you’re broke, you can use a big storefront window that you can look at while your lard ass walks by. Nothing makes your tired old boring home gym more interactive than putting up a large screen that looks just like a BIG MIRROR, because it actually IS a big mirror. You can decorate your big mirror with your own personalized motivational reminder-PostIts stuck to the places you know you’ll notice because you know your vain ass and how you going to look at yourself in the mirror and notice that reminder-Post-It and find it compelling again … some people [like farmers or steelworkers who lift heavy stuff all day] might be tempted to think this big mirror thing is bit of a joke fitness-wise, but people really do spend ginormously unbelievable serious amounts of money on the INDUSTRY known as “HOME workout mirrors”.
Apple Fitness Accessories: It’s way more than just subscription service with on-demand workout videos and guidance by Apple Watch … possibly the ONLY thing worse than celebrity fitness app horsehockey or the Peloton clotheslines, would be stinking iOS road Apples that have piled up everywhere on the info superhighway in the last 25 years … Apple product’s are far WORSE addict-wise than Gillette razor blades ever were that were sold to suckers 100 years ago who needed to shave to look like they had been WWI Vet-heroes who got in the habit of shaving to make sure the gas masks fit … the Gillette design or advertisements had half of a purpose, ie if you have ever been through a pepper spray or gas training in the military or corrections or police work, you understand why gas masks need to fit … but the problem with the Apple fritterware from the mock turtleneck crowd is their fashion statement cuts deeper.
FitOn, like HardOn, except for hard workouts… free, as in free beer, free workout videos across various disciplines with live and on-demand options [with cookies and a tracking app with ads]… is that a FitOn app or are you just happy to see me? As you might guess from an app clearly named to mine the erectile dysfunction comedy vein, the workouts practically write themselves … but, no, this is typical and NOT one bit funny.
Additional resources:
Nike Valiant Labs: A new internal venture incubation arm of Nike, focused on building businesses that serve consumers in new and unimagined ways, but there are probably OTHER startups incubators in the health/fitness/medical/insurance industries for this kind of thing … because “Just Do It” swooshes do not really get it done for most people … what might actually generate revenue for a startup product would be something developed to go beyond just nudging, with verification assistance to force the already DOUBLY-MOTIVATED user to comply, ie someone who HAS TO change behavior, OR ELSE [and, just to be sure, somebody in authority’s going to be checking to make sure] … we can think about better adherence to a person’s own VERY PERSONAL prescribed holistic treatment program, like Sweetch … the whole exercise BECOMES understandably PERSONAL when there’s at least something like a medical professional involved OR the person who has change behavior is court-ordered to do so OR someone is at least incentivized by an employer-driven quasi-mandatory health improvement regimen … AI could crossover into PERSONAL territory really quick if there’s a PERSON of authority who’s going to be using AI to check and drive accountability.
Crunchbase: Search for companies categorized as “artificial intelligence” and “fitness” to find relevant ventures … such as ArtiFit AI Personal Trainer or Volt Athletics
Bard search: Using specific keywords like “AI personal trainer” or “personalized fitness app” to discover various options.
ProductHunt: Search for “fitness” or “personal trainer” to find relevant products and services … or just browse to find new and better ideas for lots of different things … why people follow the news or sports, when they could find something interesting on the Trending Topics page on ProductHunt is beyond me.
GitHub open source repositories: Search for recentish “personal AI trainer” repository with at least a couple forks, such as AasanAI or MindfulMove or MyCoach. The ONE thing that would be better about an open source app is that the process of developing such an app would be more transparent, but it would force to developer to REALLY THINK about what they are coding and WHY anyone would care … and, if they really got serious about doing that, some developer would probably write a post pretty much exactly the one you’re reading right now.
Giving up and asking AI for help:
At some point, we were kind of out of ideas … so we help generate and then copied the following prompt from Surfsite.AI and pasted our prompt into the usual suspects, such as Bard or NVIDIA’s Llama 2 70:
> [Instructions]
> Help me find open source development communities, apps, new ventures, or startup incubators. They should be focused on providing a "Personal AI trainer" or a coaching tool. The goal is to provide better adherence to custom treatment programs. Limit your responses to open-source development communities, apps, new ventures, or startup incubators that specifically focus on Personal AI trainer or a coaching tool for custom treatment programs. When suggesting resources, include details about their features, user reviews, and any relevant success stories. Be transparent about any limitations or potential biases in your responses, and clearly indicate if you are unable to provide information on certain aspects. Ensure that your responses align with industry best practices and ethical considerations related to personal AI trainers or coaching tools for custom treatment programs.
> [Context]
> Consider the challenges and obstacles that users might face when using personal AI trainers or coaching tools for custom treatment programs, and suggest ways to overcome these challenges.
> [Audience]
> Assume that I have a basic understanding of personal AI trainers and coaching tools, but I am looking for more advanced or specialized resources.
> [Output]
> Use a friendly and conversational tone in your responses, as if you are assisting a colleague or friend. Include the following keywords in your responses: open-source, personal AI trainer, coaching tool, custom treatment program, development community, app, new venture, and startup incubator. Keep your responses concise and to the point, focusing on the most important information.
The answer that we got back was:
> I'm happy to help you find open-source development communities, apps, new ventures, or startup incubators that focus on personal AI trainers or coaching tools for custom treatment programs.
One great resource is the Open Personal AI Trainer (OPAT) community, which is an open-source platform for developing and sharing personal AI trainers. OPAT provides a range of tools and resources for creating custom treatment programs, including natural language processing, machine learning, and data analytics. The community is active and supportive, with a strong focus on collaboration and sharing knowledge.
Another useful resource is the Coaching Toolkit, which is an open-source library of coaching tools and techniques. The toolkit includes a range of coaching methods, such as cognitive-behavioral therapy and solution-focused coaching, and provides a customizable framework for creating personalized coaching plans. The toolkit is designed to be flexible and adaptable, allowing coaches and trainers to tailor their approach to the needs of their clients.
If you're looking for a more comprehensive solution, you might consider the AI-powered coaching platform, LifeX. LifeX is a new venture that uses machine learning and natural language processing to provide personalized coaching and support. The platform offers a range of features, including goal-setting, habit tracking, and mood monitoring, and provides users with personalized recommendations and guidance. LifeX has received positive reviews for its effectiveness and user-friendly interface.
Finally, if you're interested in exploring startup incubators, you might consider the AI for Good accelerator program. This program is focused on supporting startups that use AI to address social and environmental challenges, including health and wellness. The program provides mentorship, funding, and resources to help startups develop and scale their solutions.
In terms of overcoming challenges, one potential obstacle is ensuring that personal AI trainers and coaching tools are tailored to the specific needs and goals of each individual. To address this, it's important to invest in user research and testing to ensure that the tool is effective and user-friendly. Additionally, it's important to consider ethical considerations, such as data privacy and security, and to ensure that the tool is transparent and explainable.
I hope these resources are helpful in your search for open-source development communities, apps, new ventures, or startup incubators focused on personal AI trainers or coaching tools for custom treatment programs. If you have any further questions, please don't hesitate to ask!
Unfortunately, the AI response was typical of the state-of-the-art AI hallucination-ware dishing up nebulous cliche answers and bogus bloviation.
As one might guess, before actually trying to make sense of this answer … there just isn’t enough information available online readily to definitively tell you anything something like Open Personal AI Trainer (OPAT) community. It’s EITHER a new or niche community potentially abusing a less-than-common acronym about Out Patient Antimicrobial Therapy or something similarly inapplicable OR it’s complete rabbithole hallucination of the kind we have come to expect from generative AI.
Although the generative AI response was not especially useful in a direct sense, attempt to understand where there might be some tangentially related answers did ultimately furnish an extremely helpful page from the Institute of Personal Trainers on AI tools for personal trainers … and this does sort of remind us, once again, of the EXTREMELY IMPORTANT part of campaigns and development efforts that is often overlooked: IF anything meaninful is going to be legitimately effective, it is going to NEED TO BE PERSONAL … and that means something like incentivized structure to drive daily/weekly accountability to an authority figure.
AI just ain’t gonna cut it on really changing the game of making people more fit … it’s just one more fad … AI is really no different than expecting to fit, young and famous by looking at a 40-year-old a Jane Fonda VHS tape you just got from Goodwill … or otherwise, AI is no different than exercise equipment that get’s used as clothes drying rack, but it might have an appealing structure to wonks into matrix algebra … or otherwise, AI is no different than a $20 fitness tracker watch pedometer that somehow got marked up a another expensive, state-of-the-art $200 FitBit or $400 iWatch quantified self fashion accessory …
The PROBLEM with the whole “The Singularity Is Near” crowd is that AI is fundamentally impersonal and INHERENTLY inhuman. So just like VR sex, there’s not going to be any PERSONAL accountability or bonding or reproduction that happens … but the delusions will persist, because AI fanboys have invested a lot of emotional capital in AGI, but AI will FAIL.
AI has it legit uses … AS a tool, but AI that’s expected to do more will always fail because this personal accountability flaw that’s inherent in AI. Most AI, especially AI that promises too much, is still just a gimmick and always will continue to be UNTIL people understand that AI is like a pocket calculator … a nifty pocket calculator can do math tricks like “Multiply 8346 X 2371” faster than most humans or even faster than electromechanical adding machines, but that does mean calculator can manage a business or even do a great job every day tallying up the number eggs your chickens laid.