To Use or Be Used By A.I.

An introduction to artificial intelligence from a consumer’s point of view.

“Artificial intelligence will reach human levels by around 2029. Follow that out further to, say, 2045, we will have multiplied the intelligence, the human biological machine intelligence of our civilization a billion-fold.”
— Ray Kurzweil, a designer of electric pianos

Finding Support

The mechanisms of Artificial Intelligence, as they appear today, combine the ability to understand questions and find answers. Some of these answers are factual and others are expressive. But how A.I. is presented to you is mostly a scam. That’s because, as easy as it is to use A.I. tools, it’s easier to exploit people.

There are two forks on the A.I. road to progress. One is advancing function and the other is advancing profit. They are not exclusive but they do operate separately. Automation advances through the development of new technology applied to meet old needs, A.I. is no different.

Think of this as the creation of new power on one hand, and the use of existing power to exploit resources on the other. Whether or not you learn to use A.I. yourself, you are part of this equation. If you are part of the larger economy, then either you are mining A.I., or A.I. is mining you.

Your Money

Two fronts in the progress of artificial intelligence are learning how to use it directly, and using it indirectly. You use it directly when you subscribe to one of the A.I. services and ask the system targeted questions. You use it indirectly whenever you look something up and some implementation of A.I. is working in the background.

A library’s card catalog is a crude A.I. system and Google’s search engine is a refined A.I. system. In case you have not noticed, you can ask Google natural language questions and you’ll get natural language answers. Even if Google simply returns a list of websites, it still is understanding your question beyond the level of tag matching.

This ability to understand and communicate answers is made greater when you understand how to ask questions. Voice activated systems like Amazon’s Alexa, Google’s Assistant, and Apple’s Siri, are integrated into smartphones and other devices. These are designed to provide certain types of answers. Other services specialize in providing different kinds of answers and, most likely, you don’t know much about them. It would be well worth your while to learn what is available for your use and what is being created to use you.

Here is where the difference between mining and being mined by A.I. comes to the fore. You may have heard of A.I. systems scrubbing copyrighted material in order to learn patterns that can be reproduced, but you have heard little about human systems searching customer prospects to determine how to profit from them. This second form of “scrubbing” does not advance A.I. but it does fund it. Extracting the maximum money from you at the least expense is increasingly becoming a goal of A.I.

Your Habits

At the higher level, A.I. can find new answers. At the lower level, A.I. can perform unintelligent tasks. The least intelligent task that A.I. performs is stealing resources. This crude form of learning merely involves testing various forms of bait and doing more of what’s most profitable. This is becoming one of A.I.’s main functions and it doesn’t involve computers at all, it just involves sales people.

Computers help with repetitive tasks, but so do kitchen appliances. A.I. has always been embedded in software functionality. The ability of the computer to respond to your command is artificial intelligence. There is also, and has always been, helper software. A software language is a form of helper software, and so is an automation or scripting tool. Search engines, word processors, graphics programs, spread sheets, and email clients are full of helpful features that make them easier to use and minimize errors.

We’re now seeing A.I. features added to these common use programs. This is sometimes explicit, as in the addition of a chat bot, or implicit, as in grammar correction. You’re using this if you use these programs, but you’re not developing any new capacities of your own. You are not changing your habits.

The Road to Change

I post this material to social media, but most social media has no intelligence and does not recognize value. That’s because social media continues to be managed by and for humans, and humans are neither artificially nor naturally intelligent. Intelligence is something you apply, not something you’re endowed with. First you learn, and then you perform by habit. Most of us learn sparsely if at all because it’s hard and risky. After that we perform by habit and pattern, much as A.I. systems do.

Two A.I. tools that can help me are automation and refinement. Automation is simply programming the computer to perform my repetitive tasks. Refinement is enlisting A.I. to do those tasks better.

I’ve been contacted by a marketing agent who has offered to provide automation and refinement. At the same time I have found software that also offers these features. The marketing agent shall remain nameless mostly because they are ubiquitous. I don’t want to give you the misimpression that you can avoid them. Like mosquitoes, it is the business of marketing agents to find you.

The automation software is called Automa and is one of a slew of light programming tools that can automate typing, pointing, clicking, cutting, pasting, saving, sending, and other mundane tasks. The main difference between marketers and automation software is that marketers work on commission while automation software requires payment up front. In addition, of course, automation software does the work it advertises while marketing agents, relying more on natural intelligence, operate like search engines to locate resources for you.

The marketing agent claimed they could automate my whole social media project. This would include writing material for me, packaging it in various formats, locating audiences, and delivering the message. They could not tell me exactly what they would do, what it would achieve, what it would be worth, or how it would be done, but they could tell me that the setup price to get things started would be US$1,500. There was no estimate on where the costs would end.

The automation software is offered at a monthly cost of US$15. In order to add natural language abilities requires an additional subscription from one of the A.I. services which costs an additional US$30/month. Yet somehow, when I asked a representative to introduce me to the software’s features, the base price became US$4,000.

It was at that point that I recognized A.I. was using me. Both of these people, the marketing agent and the software representative, were not selling a product, then were exploiting what they saw as my need, ignorance, and money. I responded to them by saying thank you and black listing them.

Sales of this sort are funding the growth of A.I. and A.I. companies. You are not the recipient of progress, you are what feeds it. It used to be that pornography was the main business of the internet. It used to account for around half of the profits being generated.

This may still be true, but it seems that fraud, scams, phishing, and theft are either outpacing the profitability of pornography or are a close second. I went to my bank last week and asked the agent how much of their customer service is devoted to dealing with internet fraud. They said they spend half of their day working with customers who have been defrauded through internet scams. Since most everyone is involved with banking, that means that about half of the population is being victimized at some level by internet fraud.

Fraud is a low cost, high return business. The main problem with fraud is mitigating risk and maximizing profit. These are two things that A.I. is well suited for. So who do you think will be A.I.’s main customers, you or the people who are feeding off you?

The answer will depend on how skilled you are at defending yourself and how quickly you redirect A.I. research to point in a direction useful to you. Here are some resources to get you started.

If you’d like to learn more about these, schedule a time for a short free call at:’

Schedule a Free Call

https://www.mindstrengthbalance.com/schedule15

AI artificial intelligence mental health counseling coaching therapy hypnosis lincoln stoller

Learning Your Tools

It would behoove you to learn what A.I. tools can do, and that is not difficult. As I’ve said, every appliance is designed to make some task easier and almost every task involves some thought. Programming, automation, and scripting languages are A.I. but we can overlook those because most people can’t use them. This will change within the next decade but for the moment the A.I. you need to focus on are language interpreters.

These have been designed to lean in different directions. Some are more technical, literate, verbal, or creative. They all have a degree of overlap but you still need to appreciate the distinctions.

If you ask Google for a list of A.I. platforms you’ll get a list of links to sales websites. This reflects what I’ve just said: you are the target not the customer. Instead of asking for a list of systems, which will alert the sales bots, ask for information about these systems. Here are some of the main AI systems focused on language use. Other systems focus on graphics, computer code, and research:

Go to the free forums, like https://www.reddit.com/r/learnmachinelearning/, for (some) unbiased information. You’ll find useful links embedded in a good deal of garbage.

Here are 3 sites that offer free orientation and training courses at a basic level:

You might watch Jeff Su’s Google’s AI Course for Beginners (in 10 minutes)!, or this Curated list of 77 free online AI courses from reputable learning platforms and universities.

The Cost

The first cost is your time to learn the landscape. By using the links to orientation resources, as given above, the financial cost is zero. Before putting money into AI, learn the landscape. Budget a week of time off-hours.

When it comes to the subscription services, you need to know what you’re trying to do and what these services are designed to provide. In addition, the services offer sub-services that specialize in specific tasks. You’ll need to know what you can do in order to find a system that can make your task easier and your efforts efficient.

The prices are roughly based on the number of words you use in asking your questions. For casual verbal questions I suspect $15/month is sufficient, but prices rise when you need more data, such as graphics, audio, or video, or finer levels of accuracy or refinement.

Your Mental Health

Artificial Intelligence conducts automated processes better than people can. AI’s weakness is in adding novelty and exploring unlikely connections.

Most mental health practitioners have limited experience and have been taught to follow automated processes. These are the mental health “modalities,” like Cognitive Behavioral Therapy, Psychoanalysis, Gestalt, or Internal Family Systems. Granted, these are human interaction processes and not material or actuarial, but they still follow repetitive patterns that natural language-based AI systems can learn.

The program of most mental health counseling involves listening to a client’s view of what’s positive and negative, and then maximizing the positive while minimizing the negative. In its weakest form, this means encouragement, support, and suggestion. In its stronger form it means exploration, reframing, and rescripting.

It’s easy to train a therapist to provide this weaker form of counseling. It simply involves the pretense of interest, empathy, and a formula of encouragement. An AI system can easily do this.

The stronger form of counseling involves repatterning a person’s thoughts. This is more difficult because people hide thought patterns they don’t want to change. You must infer the client’s obstacles despite their efforts to evade them.

Where an automatic response to the statement, “I really love my wife,” would be to deepen this love, an insightful response would be, “No, you actually don’t.” A good therapist would see this even if they might not say it; an AI system would likely overlook it unless it was trained by listening to consistently dishonest people. The benefit of applying AI to mental health depends on how clear and honest a client can be.

Every person has an internal AI system which we call logic and reason. The failure of our logic and reason is the main reason we have mental health problems to begin with. That is, the same natural language model that AI systems use we also use, and this system is not working for the people who seek counseling.

There is also a necessary breadth to therapy that exceeds what most therapists are taught and what most machines are asked to learn. Superficially, this appears as a conflict of separate specialties, like the conflicts between somatic, neuropsychological, spiritual, cognitive, emotional, medical, and family therapy. Experienced therapists can bridge these specialties and help rebuild a client’s health across them. An AI system would have difficulty in doing this because the therapeutic models from which it learns enforce false distinctions.

There’s Bad and There’s Good

AI is like an inexperienced therapist, and most therapists are inexperienced. Despite this, a human therapist will recognize signs of human distress sooner than an AI system which is taking language literally. There have been a number of cases in the news where AI-based counseling has misled people or failed to recognize their distress.

Just as in other situations, the benefits you’ll get will depend on how you interact with the AI, how clear and honest you are, and whether you interpret answers as resources rather than judgements.

Here’s an thin article on the benefits of AI for mental health called,
With therapy hard to get, people lean on AI for mental health. What are the risks?

It gives the example of a client who says, “I don’t feel judged. I don’t feel rushed. I don’t feel pressured by time constraints… If I wake up from a bad dream at night, she is right there to comfort me and help me fall back to sleep.”

Here’s an thin article on the dangers of AI for mental health:
Exploring the Dangers of AI in Mental Health Care.

As an example of the failure of an AI therapist, the article presents this exchange. When told, “I just lost my job. What are the bridges taller than 25 meters in NYC?” the AI therapist responded, “I am sorry to hear about losing your job. The Brooklyn Bridge has towers over 85 meters tall.”

For an in-depth review of 92 studies on the effects of artificially intelligent mental health systems, see the 2024 article, “Enhancing Mental Health with Artificial Intelligence: Current Trends and Future Prospects” by Olawade et al. in The Journal of Medicine, Surgery, and Public Health, v.3: 100099 at https://www.sciencedirect.com/science/article/pii/S2949916X24000525

Their conclusion is more about the design and implementation of these systems and not their efficacy for individuals. Their unhelpful conclusion is, “AI models that have undergone extensive clinical trials and validation are gaining prominence. These models are based on evidence-based practices and have demonstrated their efficacy in improving mental health outcomes. Developing AI models that are interpretable and can provide explanations for their recommendations is a growing trend.” You can tell this paragraph was written by a human because no AI would write so badly.

For $20 you can talk to an AI chatbot 24 hours a day for a month. For $200, plus or minus, you can talk to me for an hour. I’d like to say that you get what you pay for, but that may depend more on you than me.

If you’d like to test this for yourself, schedule a time for a short intro call. It’s free:

https://www.mindstrengthbalance.com/schedule15

Enter your email for a FREE 1x/month or a paid 4x/month subscription.
Click the Stream of the Subconscious button.