Do You Think Anti-AI Apps Are A Value Skew?

[Learn To Live]

I was looking into it, and I think there can be a legitimate marketplace for it. Consumers seem to dislike apps that are "powered by AI." Additionally it seems that AI smartphone features are also unpopular. While this can be explained by issues with AI accuracy, poor execution, or people using it as a quick cash grab. I'm thinking this can work with a few groups we could serve especially well.

1. Anti-AI groups: People who are against AI due to things like it depressing wages, or leading to layoffs in their sector such as customer service.
2. Climate Activists: AI takes a ton of energy to train, so you could probably appeal to this demographic by saying how much energy the app uses compared to one with AI, and what that means for climate change.
3. People who care about privacy: Sending user data to train an AI might be something privacy focused users dislike. By marketing it to them as being private, and possibly having the source code be available it can be something that tips them off in this direction. These are probably the best market of the 4 if the app involves tech specific knowledge that their colleagues, family, and friends will ask them about.
4. People who think algorithms are better for solving specific problems: This is where I fall into. While I do think statistical models can help, I'd personally prefer a tailor made algorithm to solve the specific problem, plus it makes it easier to explain to other people.

What do you guys think?

 

Dislike ads? Become a Fastlane member: Subscribe today and surround yourself with winners and millionaire mentors, not those broke friends who only want to drink beer and play video games. :-)

[MJ DeMarco]

What do you guys think?

I hate it and don't think it's a value skew.

You can be all these things and not market position as "Anti-AI" -- that just sounds like someone who hates the automobile and wants to promote "horses and buggies" because it's always been this way!

 

[Learn To Live]

I hate it and don't think it's a value skew.

You can be all these things and not market position as "Anti-AI" -- that just sounds like someone who hates the automobile and wants to promote "horses and buggies" because it's always been this way!

Thanks for the feedback. That was the demographic I was thinking about, but I suppose marketing it that way might cause issues.

I should probably find a book on the subject of identifying markets without also isolating others.

 

[Kevin88660]

I was looking into it, and I think there can be a legitimate marketplace for it. Consumers seem to dislike apps that are "powered by AI." Additionally it seems that AI smartphone features are also unpopular. While this can be explained by issues with AI accuracy, poor execution, or people using it as a quick cash grab. I'm thinking this can work with a few groups we could serve especially well.

1. Anti-AI groups: People who are against AI due to things like it depressing wages, or leading to layoffs in their sector such as customer service.
2. Climate Activists: AI takes a ton of energy to train, so you could probably appeal to this demographic by saying how much energy the app uses compared to one with AI, and what that means for climate change.
3. People who care about privacy: Sending user data to train an AI might be something privacy focused users dislike. By marketing it to them as being private, and possibly having the source code be available it can be something that tips them off in this direction. These are probably the best market of the 4 if the app involves tech specific knowledge that their colleagues, family, and friends will ask them about.
4. People who think algorithms are better for solving specific problems: This is where I fall into. While I do think statistical models can help, I'd personally prefer a tailor made algorithm to solve the specific problem, plus it makes it easier to explain to other people.

What do you guys think?

Instead of hating AI, you can see these as opportunties for problem solving.

(1) Jobs and layoff: retraining works to learn to use AI. There are companies willing to spend money to teach their older workforce how to use chatgpt. The 65 year old yet to retire doctor or engineer…

Sam Altman is working on ubi and eye scanning technology. I know this kind of scale and investment is out of reach for most but just for ideas sharing.

(2) Climate issue : Energy efficiency in AI. If your LLM models is less reliant on raw computation strength it will be more energy efficient…out of reach for most here for such a big project but just for illustration.

(3)privacy need: privacy tools or prompt that make sure LLM don’t store your personal data. Not exactly sure how it can be done but it’s an angle.

(4) Providing services where current AI models suck at. But this is at risk for quick disruption…

 

[Learn To Live]

Thanks for your feedback. I'll look into some of these in more detail later as you've given me a lot to think about.

Instead of hating AI, you can see these as opportunties for problem solving.

I'm ambivalent to AI, I think it's useful for a specific subset of problem, but just hype outside of that. However, I hear some people complaining about it a lot. If you asked me as a programmer, things like libraries, or low-code/no-code would threaten job security more than AI. I use libraries a lot in work and personal projects when useful.

(1) Jobs and layoff: retraining works to learn to use AI. There are companies willing to spend money to teach their older workforce how to use chatgpt. The 65 year old yet to retire doctor or engineer…

This would probably be the easiest suggestion to do. The hard part would be getting people to work with it rather than against it. Perhaps giving it a wrapper with their grandchildren or something to get them to not think about it being a strange new thing.

Sam Altman is working on ubi and eye scanning technology. I know this kind of scale and investment is out of reach for most but just for ideas sharing.

Eye scanning might actually be easy for me. If it's a smartphone app then it's largely a problem of finding out what to look for in an eye, analyzing the important information, then reference it using existing tech.

Assuming by UBI you mean Universal Basic Income, I don't have the money for it. I'm also not a huge fan, but my opinion isn't important as I'm not a consumer, and I'm not super educated on the matter.

(2) Climate issue : Energy efficiency in AI. If your LLM models is less reliant on raw computation strength it will be more energy efficient…out of reach for most here for such a big project but just for illustration.

Honestly, turning the emissions into something useful might be a more profitable idea. Creating something that takes in Carbon Dioxide and produces oxygen might even be easier.

As for making an LLM that doesn't rely on as much computational strength, I would need to look into the lower bounds for computational complexity. The fastest matrix multiplication algorithm is actually an unsolved problem in Theoretical Computer Science. A similar unsolved problem that could be useful regarding encryption would be a one-way hash function, which would also imply P != NP (I'd get $1,000,000).

Figuring out which data to keep, and which to discard might be the important thing. Simply by having a lower amount of data to train on, it will reduce the size of matricies, and thus improve computation complexity. However it may lose information on outliars, but would probably work "well enough" for average situations.

(3)privacy need: privacy tools or prompt that make sure LLM don’t store your personal data. Not exactly sure how it can be done but it’s an angle.

Assuming that they're calling OpenAI, you can at best anonomyze what is said, but that might impact the output. You could distribute the model locally if you have one, there are FOSS options. In these cases you can opt-out of data collection.

In general this might be unfixable without doing something different. The thing that makes AI useful is the data it's trained on, without collecting massive amounts of data, it's useless. For example, the big 5 was created by taking the list of words used to describe people then seeing which ones basically meant the same things, eventually boiling it down into 5 (although HEXACO exists, and we think that you need more dimensions to adequately describe personality). There is a possibility to do something like that with a dictionary and using syntax to construct a system of semantics to get a good idea of a psuedo-model of English. This would require less data and training, but cost accuracy, and causes many other problems. However I'm sure those can be mitigated.

(4) Providing services where current AI models suck at. But this is at risk for quick disruption…

I don't think this is a problem, I'm talking about situations where AI is overkill, or wose than an algortihm. It's like trying to nuke a country if you want their farmland, you'll destroy the thing you were after in the first place. If you have an AI optimizing your schedule that's overkill, the best thing it can do is analyze your behavior and could readjust it based on that, but let's say it sees you keep playing video games, it does nothing that actually helps you. On the other hand, having an algorithm based on say habit stacking could work just as well, and work in expected ways 100% of the time.

Personally, it annoys me when I see ads for it and think it's a cash grab for people who don't understand the tech. Kind of like anti-virus software, my parents were paying over $100/year for things that Windows Defender does. I'm not trying to scam people for a quick buck, it's unsustainable and unethical. However, this could involve calling people and telling them about what software they don't need on their computer, and giving them alternatives, with an upsell option of training them on using more useful applications.

 

[Andy Black]

I suppose anti-AI is pro-human. Maybe focus on the pro-human part?

"Our team of human copywriters will lovingly ... blah blah blah." (Clearly I'm not a copywriter.)

 

Dislike ads? Become a Fastlane member: Subscribe today and surround yourself with winners and millionaire mentors, not those broke friends who only want to drink beer and play video games. :-)

[EngineerThis]

I suppose anti-AI is pro-human. Maybe focus on the pro-human part?

"Our team of human copywriters will lovingly ... blah blah blah." (Clearly I'm not a copywriter.)

Bingo, just like selling a product that was "Made in America" for us Americans that is

 

[Learn To Live]

I suppose anti-AI is pro-human. Maybe focus on the pro-human part?

"Our team of human copywriters will lovingly ... blah blah blah." (Clearly I'm not a copywriter.)

That's a good point, thank you.

 

[RealDreams]

would you rather be part of the ones who invested in the dot com boom (or should I rather say, internet businesses) or the ones who whined and hated those promoting it? Where are the latter today anyway?

I would bet money those hating on AI ain't going anywhere. You can't stop AI from being developed or used.

How about learning how to code and building your AI app? Lots of open-source AI models out there. If "too difficult" is what comes in your mind, then you're a loser, sorry. You'd rather hate than solve problems.

It reminds me of the anti-work crowd. People hate other people, CEOs, businesses so they never get rich. That's the wrong paradigm to have in your mind. Reddit is as trash as it has ever been. Just look at the people cheering and promoting slaughter of CEOs. Idiots.

Moreover, trying to sell to people hating capitalism is a bad idea for one simple reason, these people are more often than not poor so you'll struggle making a lot of money from them, in my opinion. You want to target people with money to spend.

I get all the ethical talk about AI being bad. But again, you can't stop it, and those hating on it are mostly virtue signalling, trying to garner internet points. And AI can definitely be used for good. You just need to think of problems to solve and use AI to solve them. Generative AI is just one subset of AI. There's also image classification and so on. I personally believe AI is the best thing that happened for wanna be entrepreneurs in the past 15 years because it opened a whole new market.

 

[Learn To Live]

would you rather be part of the ones who invested in the dot com boom or the ones who whined and hated those promoting it?

It depends. If the company had the fundamentals in tact to where it is good regardless of if it's trending, absolutely about 48% of the companies survived the crash, so the chances are really good. On the other hand, if the only thing the company stands on is hype, I'd want to stay as far away as possible. While it might be good if I can sell high, right now I'd say getting into AI has a higher likelihood of buy high and sell low in the near future, but say 10 years from today it will likely correct.

Where are the latter today anyway?

That's a good question, I'll have to look into. I wasn't alive so I can't think of any obvious examples of small business off the top of my head. Some corporations who would've had the power to weather the storm come to mind, but in some ways they have had issues due to being unable to adapt.

You can't stop AI from being developed or used.

In that case wouldn't it be better to try creating the next big thing. Machine learning has been an idea since the early computer days, but only recently became useful. Would it be better to try developing quantum computing or some other emerging technology before it's popularized therefore starting the gold rush instead of capitalizing on it.

This does actually seem like an easier idea because it's mostly limited by imagination and what hardware can do. Not to mention if I'm the 1st/only then it automatically passes Entry.

 

[100k]

I didn't quite understand whom you are serving, the problem they have and the ideal (measurable) outcome that your solution will deliver.

 

Dislike ads? Become a Fastlane member: Subscribe today and surround yourself with winners and millionaire mentors, not those broke friends who only want to drink beer and play video games. :-)

[Learn To Live]

the problem they have and the ideal (measurable) outcome that your solution will deliver.

I'm currently looking for problems that I can solve. Some previous ideas were either too small (job boards) or too big (anti-viral medication). Right now I'm trying to avoid the problem of having a solution looking for a problem.

An example of an idea I've had that might not be terrible, and have a few pages of notes on is a recipe sharing website where people can share their favorite recipes to make things from scratch, find ways to optimize prices, and possibly partner with stores to offer coupons to customers. However, I'm not sure if this is an idea worth investing it, because I'm unsure if there is a market for it, and users would be the most important business model.

whom you are serving

I'm trying to figure that out. I mentioned a few groups in the first post: privacy focused users, people who care about the environment enough to boycott something, people who are fearful of AI or bitter, people who prefer tailored solutions. There are people complaining about "dead internet theory," so that could also be a market.

My personal opinions don't particularly matter. Although apart from work I have almost no social interaction, so I might need to collect more data on people.

 

[Learn To Live]

It depends. If the company had the fundamentals in tact to where it is good regardless of if it's trending, absolutely about 48% of the companies survived the crash, so the chances are really good. On the other hand, if the only thing the company stands on is hype, I'd want to stay as far away as possible. While it might be good if I can sell high, right now I'd say getting into AI has a higher likelihood of buy high and sell low in the near future, but say 10 years from today it will likely correct.


That's a good question, I'll have to look into. I wasn't alive so I can't think of any obvious examples of small business off the top of my head. Some corporations who would've had the power to weather the storm come to mind, but in some ways they have had issues due to being unable to adapt.


In that case wouldn't it be better to try creating the next big thing. Machine learning has been an idea since the early computer days, but only recently became useful. Would it be better to try developing quantum computing or some other emerging technology before it's popularized therefore starting the gold rush instead of capitalizing on it.

This does actually seem like an easier idea because it's mostly limited by imagination and what hardware can do. Not to mention if I'm the 1st/only then it automatically passes Entry.

Adding more context because you added more information.

How about learning how to code and building your AI app?

I code very well, to the point where I got my job doing it at 20, and often have coworkers ask me for technical advice. I have zero aversion to learning something new. This year alone I've read textbooks on Real Analysis and Number Theory and have ideas as to how to use them to improve my programming. Right now I'm deciding between learning more Linear Algebra and Complex Analysis. I have zero doubts in my academic abilities, socially on the other hand, leaves a lot to be desired.

I have notes on how to combine concepts in psychology with training an AI model. However, this solves no problems. I have solutions, not problems.

you're a loser, sorry.

I'm not offended, I'm here to learn. I'm also not insecure in this area.

It reminds me of the anti-work crowd. People hate other people, CEOs, businesses so they never get rich. That's the wrong paradigm to have in your mind. Reddit is as trash as it has ever been. Just look at the people cheering and promoting slaughter of CEOs. Idiots.

Moreover, trying to sell to people hating capitalism is a bad idea for one simple reason, these people are more often than not poor so you'll struggle making a lot of money from them, in my opinion. You want to target people with money to spend.

Good point, so does that mean, I should focus on a B2B business and ignore consumers? If so, any advice on finding problems businesses are having? I've been thinking that creating models that predict stock fluctuations based on weather patterns. Right now I'm trying to learn as much as possible to improve the quality of my ideas as well as speed on executing said ideas. When studying differentiability criteria I was thinking about ways to recreate crime scenes.

I get all the ethical talk about AI being bad. But again, you can't stop it, and those hating on it are mostly virtue signalling, trying to garner internet points. And AI can definitely be used for good. You just need to think of problems to solve and use AI to solve them. Generative AI is just one subset of AI. There's also image classification and so on. I personally believe AI is the best thing that happened for wanna be entrepreneurs in the past 15 years because it opened a whole new market.

I agree image classification is a great use for machine learning, this is what I was referring to when I said statistical models can help.

Also, if you think AI is the best thing to happen in the past 15 years, would you say it's a good idea to develop it further rather than solve problems with it? I am confident in my ability to make it better, and would even have fun doing it, but I'm unsure if it's profitable.

 

[100k]

I'm currently looking for problems that I can solve. Some previous ideas were either too small (job boards) or too big (anti-viral medication). Right now I'm trying to avoid the problem of having a solution looking for a problem.

An example of an idea I've had that might not be terrible, and have a few pages of notes on is a recipe sharing website where people can share their favorite recipes to make things from scratch, find ways to optimize prices, and possibly partner with stores to offer coupons to customers. However, I'm not sure if this is an idea worth investing it, because I'm unsure if there is a market for it, and users would be the most important business model.


I'm trying to figure that out. I mentioned a few groups in the first post: privacy focused users, people who care about the environment enough to boycott something, people who are fearful of AI or bitter, people who prefer tailored solutions. There are people complaining about "dead internet theory," so that could also be a market.

My personal opinions don't particularly matter. Although apart from work I have almost no social interaction, so I might need to collect more data on people.

Personally not too keen on the recipe idea.
Sounds more like a community on discord or skool ... Where foodies can share their ideas, not sure what the economic value is there. Seems like a hobby.

You're too broad and generic with your targeting.

Who exactly cares about privacy? B2B business whom it's crucial to have unhackable security like PayPal, stripe, GoDaddy, MailCheat(Chimp)?

High net worth individuals with crypto accounts worth millions?

You have to be way more niche and targeted.

People that care about the environment to the point of protesting... You mean broke college kids or hipsters? How will you monetize them?

Maybe you mean farmers that own at least 100 acres and 1 degree change in the climate can cost them millions.

Be more specific and you'll be able to reach out to those people and then talk to them about their problems and then come up with a solution.

 

[RealDreams]

Adding more context because you added more information.


I code very well, to the point where I got my job doing it at 20, and often have coworkers ask me for technical advice. I have zero aversion to learning something new. This year alone I've read textbooks on Real Analysis and Number Theory and have ideas as to how to use them to improve my programming. Right now I'm deciding between learning more Linear Algebra and Complex Analysis. I have zero doubts in my academic abilities, socially on the other hand, leaves a lot to be desired.

I have notes on how to combine concepts in psychology with training an AI model. However, this solves no problems. I have solutions, not problems.


I'm not offended, I'm here to learn. I'm also not insecure in this area.


Good point, so does that mean, I should focus on a B2B business and ignore consumers? If so, any advice on finding problems businesses are having? I've been thinking that creating models that predict stock fluctuations based on weather patterns. Right now I'm trying to learn as much as possible to improve the quality of my ideas as well as speed on executing said ideas. When studying differentiability criteria I was thinking about ways to recreate crime scenes.


I agree image classification is a great use for machine learning, this is what I was referring to when I said statistical models can help.

Also, if you think AI is the best thing to happen in the past 15 years, would you say it's a good idea to develop it further rather than solve problems with it? I am confident in my ability to make it better, and would even have fun doing it, but I'm unsure if it's profitable.

You can develop AI models yourself, as long as you remain a businessman foremost and an engineer second. I see so many engineers releasing their work for free (open source), and while it helps the community as a whole, it doesn't help them financially speaking. If you can code I'd definitely focus on using AI to your advantage instead of joining and serving the anti-work woke crowd.

 

[Bekit]

People who care about privacy: Sending user data to train an AI might be something privacy focused users dislike. By marketing it to them as being private, and possibly having the source code be available it can be something that tips them off in this direction. These are probably the best market of the 4 if the app involves tech specific knowledge that their colleagues, family, and friends will ask them about.

I think an Anti-AI app is totally a value skew.

Just today, Gmail informed me that Gemini is now turned on. I'm seemingly not able to opt out.

I don't want the content of all my business emails and Google Drive files being ingested into your LLM, thanks. Same thing with Asana. All our tasks are being looked at by the Asana AI so that it can "help us" by providing AI summaries to tasks.

Soon, there's going to be no such thing as "Proprietary business intel" or knowledge assets that are private to your business.

A way to "AI-proof" your business secrets might be a nice option.

 

Dislike ads? Become a Fastlane member: Subscribe today and surround yourself with winners and millionaire mentors, not those broke friends who only want to drink beer and play video games. :-)

[Learn To Live]

Personally not too keen on the recipe idea.
Sounds more like a community on discord or skool ... Where foodies can share their ideas, not sure what the economic value is there. Seems like a hobby.

You're too broad and generic with your targeting.

Who exactly cares about privacy? B2B business whom it's crucial to have unhackable security like PayPal, stripe, GoDaddy, MailCheat(Chimp)?

High net worth individuals with crypto accounts worth millions?

You have to be way more niche and targeted.

People that care about the environment to the point of protesting... You mean broke college kids or hipsters? How will you monetize them?

Maybe you mean farmers that own at least 100 acres and 1 degree change in the climate can cost them millions.

Be more specific and you'll be able to reach out to those people and then talk to them about their problems and then come up with a solution.

Thank you for the more advice on narrowing down a niche, I'll keep this in mind. I'm guessing it would be a good idea to think about this until I know say an exact person to ask.

Example of scoping in: Environment > Farmers > Farmers with lots of land > Farmers with lots of land in specific regions most impacted.

From there would it be wise to cold call, or buy something from them and mention that I have a product that can help them?

 

[Learn To Live]

I think an Anti-AI app is totally a value skew.

Just today, Gmail informed me that Gemini is now turned on. I'm seemingly not able to opt out.

I don't want the content of all my business emails and Google Drive files being ingested into your LLM, thanks. Same thing with Asana. All our tasks are being looked at by the Asana AI so that it can "help us" by providing AI summaries to tasks.

Soon, there's going to be no such thing as "Proprietary business intel" or knowledge assets that are private to your business.

A way to "AI-proof" your business secrets might be a nice option.

That's a very good idea, corporate espionage can be a real risk. In that case it might be wise to recreate an office suite and market it that way.

I'm wondering if doing an indirect pentest using things like Gemini for passive information gathering would be a good way of saying "hey, Google leaked your confidential records, but you can use my product for $100/user for life." I'd need to look into it more, but I've considered pentesting and having a money back guarantee if I find no critical vulnerabilities within 72 hours (time frame is not final). It fails Time, but it would be something I have no problem doing, not to mention if I brute force my way to say 5 milion over 5 years I won't be complaining, and I'd probably still run the business.

 

[AFMKelvin]

I personally think there's a niche in this somewhere for undigital lifestyle. People that don't want more digital gadgets. For example I wish they sold an analog car. A car without a computer running things. The only thing you would need is a battery for the headlights and radio. Also whenever I want to buy something online I start by googling the name of what I want followed by analog. One category that is good at keeping it analog is the beauty category. You can find electric razors, nose hair trimmers etc. I managed to find double edge razors that don't need to be recharged to shave. And scissors instead of electric nose hair trimmers. I like the peace of mind that I didn't forget to charge something. Or that a transitor got busted now the item doesn't work. I think you might be on to something and hopefully someday there's a undigital store where everything inside the store is chipless.

 

[kiwiana]

Have you dug into Anti-AI [what?]? AI is a lot of things. Anti-AI itself could mean 1000s of things. For example, I heard recently about a company that is combating the hallucinations in gen AI by using a different kind of AI (sorry, I forgot the term) so that it will create a knowledge base of a company's internal documents and when a user queries asking for something, it will send a link to the document. It doesn't hallucinate because it doesn't generate new documents. It takes all the docs a company has and ingests it, then it can provide a path to someone to find what they're looking for. The gen AI component is still there in the chatbot interface though. It's a big timesaver for big companies that have documentation everywhere.

What other ways does AI have shortcomings that an anti-AI-shortcoming focused product (yuck, sorry terrible phrase) on fixing that problem?

 

[Devilery]

The concept of "Anti-AI" seems quite broad. Anti-what exactly? AI text? AI creatives? AI support agents? These are just some of the basic use cases, there's data processing (e.g. in the financial sector) and manufacturing, and probably a lot I don't even know about.

I'm sure there are some use cases, but personally, I wouldn't pursue it. It's not far from being anti-Internet in the 90s.

Like it or not, AI is the future, a very near future (or simply present already?). Instead of fighting against it, figure out how to ensure AI supports and complements, not replaces.

 

Dislike ads? Become a Fastlane member: Subscribe today and surround yourself with winners and millionaire mentors, not those broke friends who only want to drink beer and play video games. :-)

[100k]

Thank you for the more advice on narrowing down a niche, I'll keep this in mind. I'm guessing it would be a good idea to think about this until I know say an exact person to ask.

Example of scoping in: Environment > Farmers > Farmers with lots of land > Farmers with lots of land in specific regions most impacted.

From there would it be wise to cold call, or buy something from them and mention that I have a product that can help them?

It's too early to think about trying to sell them anything. Try to understand their problems, try to figure out what they have done to overcome the problem, figure out how much it would cost them if they didn't find a solution... Then go and try to find a solution or make one for them... Then give it away for a steep discount for the first few clients, build case studies, get more valuable insights and data from them.... Then when you have found product market fit and you know exactly whom you are helping and would get the most value out of your solution... That's when you approach them and offer to educate them about how you helped someone like them with the exact same problem and then see if they'd like to jump on a call and see if you could help them solve the same problem and save them their farm or millions of dollars within the next X years.

 

[Johnny boy]

This should help you understand what you sound like right now and how much of a bad take you have on this.

I changed AI to "the internet".

***

I was looking into it, and I think there can be a legitimate marketplace for it. Consumers seem to dislike things that are "powered by the internet." Additionally it seems that internet features are also unpopular. While this can be explained by issues with internet accuracy, poor execution, or people using it as a quick cash grab. I'm thinking this can work with a few groups we could serve especially well.

1. Anti-internet groups: People who are against the internet due to things like it depressing wages, or leading to layoffs in their sector such as customer service.
2. Climate Activists: The internet takes a ton of energy to train, so you could probably appeal to this demographic by saying how much energy it uses compared to one with internet, and what that means for climate change.
3. People who care about privacy: Sending user data to the internet might be something privacy focused users dislike. By marketing it to them as being private, and possibly having the source code be available it can be something that tips them off in this direction. These are probably the best market of the 4 if it involves tech specific knowledge that their colleagues, family, and friends will ask them about.
4. People who think pen and paper are better for solving specific problems: This is where I fall into. While I do think the internet can help, I'd personally prefer a tailor made handwritten soluton to solve the specific problem, plus it makes it easier to explain to other people.

What do you guys think?
***


It's dumb because it's so incredibly broad, generic, wrong, etc.

When the only thing you've ever learned about business is "solve problems bro", you get stuff like this.

You also get ideas like "It's a dating app but instead of matching you up with people you don't like it matches you with people you do like".

Or "It's called Helpr. You tell it things you need help with and I come and help you with it. It's going to be Uber but incredibly vague and even less profitable"

 

[heavy_industry]

Being "anti-AI" is like being anti-electricity or anti-combustion engine.

Once a new technology hits the consumer market and the genie is out of the bottle, nothing can put it back inside. It doesn't matter how hard you scream at it, or how many people show up with their torches and pitchforks.

You either adapt to the new reality or get left behind.


With that being said, "talk to a real human" or "made by real people" will become massive value skews over the coming decades, but that is markedly different from being "anti-AI".

 

[Learn To Live]

When the only thing you've ever learned about business is "solve problems bro", you get stuff like this.

Thanks for the feedback. What would you recommend learning? I'm planning on adding a comment on a list of things I'll want to look into learning.

 

[Learn To Live]

With that being said, "talk to a real human" or "made by real people" will become massive value skews over the coming decades, but that is markedly different from being "anti-AI".

That's a much better way of phrasing it, thank you.

 

Dislike ads? Become a Fastlane member: Subscribe today and surround yourself with winners and millionaire mentors, not those broke friends who only want to drink beer and play video games. :-)

[Learn To Live]

The concept of "Anti-AI" seems quite broad. Anti-what exactly? AI text? AI creatives? AI support agents? These are just some of the basic use cases, there's data processing (e.g. in the financial sector) and manufacturing, and probably a lot I don't even know about.

I'm sure there are some use cases, but personally, I wouldn't pursue it. It's not far from being anti-Internet in the 90s.

Like it or not, AI is the future, a very near future (or simply present already?). Instead of fighting against it, figure out how to ensure AI supports and complements, not replaces.

That's a good question. I'm don't know what's specifically important to customers, so I'd assume avoiding it in as many ways as possible would be what's important. However I'm probably wrong. I wouldn't consider things like data analysis such as stock trends to be in this area, nor something that most people would care about. Perhaps studying the anxieties of individuals would give me a better idea.

As for being anti-internet, there is a market for it, niche yes, but existential. There are even people who nowadays are preferring to use dumb phones.

I'm largely ambivalent to AI. I think it's a cool toy, and can be useful for certain applications where the situations are predictable; I don't think the tech is there for unpredictable situations, but it can find patterns we weren't aware of yet. I've found AI tools in IDEs to be more annoying than helpful. I'm not saying that I think they're useless, but I've found it giving me bad code snippets more than it's helped. And usually by the time I could come up with the prompt for ChatGPT or something for the code I want, I'd be faster writing it myself. One interesting use I've heard of some people doing is using it to speed up reading documentation, which sounds like a good use case.

It's worth noting I'm 22 and have 2.5 years of professional experience, so I can be missing a lot of the higher-level non-technical details. Also, I think for me having a solid foundation is more important than using the latest and greatest. Not because it's not useful, but because outsourcing my abilities without a foundation to stand upon is dangerous. Similar to how outsourcing thinking is dangerous, if I did I might have struggled more to find my first job if I only listened to my parents and focused exclusively on school, and didn't spend basically 14 hours/day learning to code for 2 years.

 

[Learn To Live]

It's too early to think about trying to sell them anything. Try to understand their problems, try to figure out what they have done to overcome the problem, figure out how much it would cost them if they didn't find a solution... Then go and try to find a solution or make one for them... Then give it away for a steep discount for the first few clients, build case studies, get more valuable insights and data from them.... Then when you have found product market fit and you know exactly whom you are helping and would get the most value out of your solution... That's when you approach them and offer to educate them about how you helped someone like them with the exact same problem and then see if they'd like to jump on a call and see if you could help them solve the same problem and save them their farm or millions of dollars within the next X years.

Thanks for the advice on how to sell once I have a product-market fit.

I agree it's too early to sell. Right now, I have no idea what problems are even worth solving, so I'm trying to figure that out. Largely I'm speculating because outside of work, occasional calls from family, and occasional giving advice to old friends I have almost no social contact. I'm too young and inexperienced to be a life coach people would take seriously.

I could probably do security based software development because breaking someone's servers indicates an obvious flaw, then I can sell a solution. It's a service so it fails Time, but I'm thinking might be a smarter idea to pursue to get more socialized and encounter more problems.

Do you have any suggestions for finding problems, such as either reading books on the topic, or interacting with people more?

I've read online that restaurants have problems with the menu. So something that processes orders and inventory changes to determine what needs to be bought more or less, and what items might be able to be removed from the menu could be a good idea if they aren't doing it already. This is an area where a statistical can be useful because it's providing insight rather than a solution.

 

[Learn To Live]

What other ways does AI have shortcomings that an anti-AI-shortcoming focused product (yuck, sorry terrible phrase) on fixing that problem?

I admit it's terrible phrasing.

I would say that if AI is the main solution it's generally going to be solve the problem poorly. For example code generated by LLMs tends to cause a lot of bugs, I would never trust code produced by any tools, so I'll probably more of a code reviewer than producer, but for code I understand less due to not producing it. Machine Learning in general such as Linear Regression and other statistical modeling techniques are useful for insight, and description, but not perscription.

I would say the best areas for it are when it's used for the purposes of gaining insights as opposed to actually doing something of practical value. For example I wouldn't trust an AI to make my schedule, but I would trust an AI to say that I tend to fall short in say reading fiction. If I have a goal to read maybe 2 fiction books and the AI allocates more time for fiction, then that can cause issues because I might not see it as a priority. If the AI allocates 0 time to it, then it can cause issues because my lack of discipline isn't really something that should mean to give up.

What it would be good for in this use case would be making a suggestion, to put it somewhere in my schedule before an enjoyable task like studying math, but there is no reason why you would need an AI to do this. It might make the application slower in general due to need to do the unnecessary processing. It's like using a tank to stop a home invasion, it's not helpful for the problem in question, and is largely expensive and counterproductive.

An example of a good use case is as you mentioned reading documetnation.

 

[Learn To Live]

I personally think there's a niche in this somewhere for undigital lifestyle. People that don't want more digital gadgets. For example I wish they sold an analog car. A car without a computer running things. The only thing you would need is a battery for the headlights and radio. Also whenever I want to buy something online I start by googling the name of what I want followed by analog. One category that is good at keeping it analog is the beauty category. You can find electric razors, nose hair trimmers etc. I managed to find double edge razors that don't need to be recharged to shave. And scissors instead of electric nose hair trimmers. I like the peace of mind that I didn't forget to charge something. Or that a transitor got busted now the item doesn't work. I think you might be on to something and hopefully someday there's a undigital store where everything inside the store is chipless.

Thank you. These are all very good points. I also think that in many ways these things also interfere with the Right To Repair, which is personally a big reason why I prefer custom desktops to laptops. I would spend more money to have something be more easily fixable than to have to buy a new one.

Also people using security cameras as a botnet, but I don't think most people care. I do think enough people care and that they'd be the people who get asked for recomendations for tech, or to set it up. For example appealing to a computer enthusiast means that you might indirectly influence their friends and family on most chioces in that domain. This is assuming it's good enough and doesn't have problems, such as historically Linux with gaming and MS office.