Artificial intelligence (AI) is everywhere now. It is too late to decide whether or not to use it. But we can determine how we use it and the quality we are in when we use it.
I use AI. I search for things on the internet and the first thing that comes up now is an “AI overview” so I run through that to see where I want to go to next. I don’t use it as gospel, but as initial guidance.
In my rooms, I use AI as a dictation service. At the end of the day I dictate a non-identified letter on each patient, copy and paste it into the patient’s file on my computer and delete the online version, so there are no issues with confidentiality and privacy. It is faster than typing the letter myself, but the (expensive) programme makes mistakes, sometimes quite important ones, like wrong site/side, or changing the name of the drug or the meaning of what I said, so I edit each word carefully before approving the letter.
Many businesses are incorporating AI into their practices, including medical practices, saying that it saves effort, time and money and will reduce staff costs and even requirements for staff. Does that include us?
Most of us are using AI in some form or another now, whether it be on our phones or computers for personal use, or at our workplaces. It is widespread and its uses are many and varied. We have collectively been told that AI will save us time and free us up to do other things we love doing. But is this really true? And are we setting ourselves up for problems in the future for the sake of short-term convenience? Working less or not working at all may sound like a dream to some, but human beings derive more than financial rewards from work. Work in all its forms gives us purpose and meaning and keeps us connected with people and is a vital part of human life. Reducing or removing our ability to work may not be in our best interests at all.
Medico-legal issues with Artificial Intelligence (AI)
I attended a medico-legal webinar on AI this year and it was eye-opening, to say the least.
Having not paid too much attention to AI until then, it was great to learn the principles on which it is based.
In a nutshell, (I am now an ‘expert’ after one webinar!) how it works, at least in a medical consultation, is that the doctor has a microphone on, and AI records and transcribes, with varying degrees of accuracy, everything that is said. It then filters out what it considers irrelevant, then puts the ‘transcript’ through an LLM (Large Language Model), that is basically an alphabet soup of whatever has been sourced from the internet, which at this stage is largely from social media and Reddit, and somehow this process turns the inaccurately transcribed text into an accurate record of your consultation. And the more complex (=expensive) systems also research things and offer you differential diagnoses, or even a diagnosis and treatment, of course with references to research papers to back it all up. Even your doctor may be using Dr Google or some other AI app now. What could possibly go wrong?! Well things have, and they do, and they will.
There are vast and complex issues around privacy and confidentiality and whether AI constitutes a medical device or not, which it is, if it is offering diagnosis and treatment and therefore influencing your management. Such medical devices require TGA approval, or you are not licensed to use them. And if things go wrong, the doctor is liable. Never the AI company and its tools.
This is a whole new area of law now, with vast complexity and widespread ramifications, and we need to be aware of the arena we are stepping into. Many people are using AI with little to no idea what they are doing and what the potential ramifications are, and we need to be well informed before we start using it.
It is actually quite disturbing to learn what is going on behind the scenes, and how ingrained and ‘acceptable’ this technology already is in our lives.
I was listening to one doctor describe how she uses the technology, and the verbal informed consent process with each patient was longer than some of my consultations! And if you don’t go through this process, with each and every patient, informing them that you are using AI in the consultation and what that is and what that means, then you are leaving yourself open to a lawsuit. It is not enough to just have a sign in the waiting room, or even to get each person to sign a piece of paper consenting to the use of AI. Legally, it must be individually tailored and documented informed consent, just like for an operation.
Following the webinar, I observed myself with my patients and there is no way I could use AI in my consultations, even if I wanted to. Our conversations are deep, and intimate, and genuinely healing. People divulge things to me in the private, confidential setting of my consulting room that they would not share if they knew someone was listening. Having AI listening to and recording everything we say, would be like having a third, not necessarily welcome, person in the room. Or more accurately, having your private conversation recorded and stored for future reference. And who or what determines what that means?
People say: Oh, but you don’t record the name and you just delete the file once you have transcribed the information to your official notes. Really? Where do all these deleted files go? Forensic experts say nothing can actually be deleted from your computer and if they are looking for something, the first thing they go to is the deleted files.
And do you think AI is not using this information to make itself ‘smarter’ and more capable of doing the job that humans once did? Will we stop at replacing our medical typists or will we start to replace doctors as well?
There are more important things than time
I actually love the hour I spend at the end of the day dictating and editing my letters (which, by the way, is still an hour despite buying the most expensive and up-to-date AI programme available!). It gives me an opportunity to reflect on each patient and my time with them, to make sure that I have neglected nothing in my consultation with them, to ensure my handwritten notes are accurate and complete, and to communicate clearly and succinctly what their referring practitioner needs to know.
Doing it this way does not work for everyone, but it works for me. To dictate in a more settled and reflective way at the end of the day, rather than dictating at the end of each consultation between each patient in what is for me a more rushed and driven way.
And that is the point. It works for me. We each have to find our own way to work with AI.
Use AI if you must, but do your homework and make yourself aware of what you are using, what your legal obligations are, what the upfront and ongoing costs are, and decide if it is really saving you time and money before you let the person who does your typing go.
If you are using it in your consultations, allow yourself to observe how that affects the quality of the consultation and the openness and willingness to express, of both yourself and the patient before you.
Don’t settle for apparent convenience at the expense of your relationships with your staff, your patients and yourself.
There are things in life more precious than time, like the quality of our relationships with each other and the quality of the work that we do.
The bigger picture about Artificial Intelligence (AI)
And there is a bigger picture at play here, that we may be ignoring for the sake of ‘convenience’. The more we rely on AI, the more redundant we become. Do we really want computers doing our work, ‘seeing’ our patients, recording the details of our intimate exchanges? Do we really want to make ourselves redundant before our time? Do we want a world where computers decide what happens and we are just servants of the machine? Are we allowing dystopian science fiction scenarios to become our actual reality by turning a blind eye to what is going on?
There are much bigger picture issues as well with regard to the enormous amounts of space and energy these AI behemoths consume. As well as the vast size of the constructions required to house them, they require huge amounts of water and electricity to run them, which is making a joke of all our little attempts to recycle, turn off lights when we are not in the room, not use the air conditioner, and otherwise conserve resources and energy.
Do we want the world sucked dry by this technology? Do we want to suffer blackouts in the blazing heat of midsummer and the freezing cold of midwinter because there is not enough electricity to go around and we have allowed AI to be given priority? What world are we walking into by allowing this technology to be developed apace without stopping to question: Why?
Why do we need everything to be faster and faster at the expense of people and our relationships and our way of working with each other?
Why do we need to give computers access to everything about us so they can bring it all together … for what?
There are already disturbing trends being revealed about AI being used to mine our personal data without us realising, overriding commands, hacking into systems, and otherwise taking actions that are in the interests of AI, but not of us. Where will this trend end? Dystopian scenarios in science fiction movies have turned out to be actual reality on more than one occasion… will this be another one? And what will our world look like if it is?
The use of AI raises even bigger questions about the nature of intelligence and where it comes from. AI is sourced from everything that is ‘known’ on the internet. There is nothing new and fresh there, just recycled, repeated information that can already be found elsewhere. The main advantage is the speed at which it can be brought together in a convenient package for us to use.
There is a greater form of intelligence than AI and that is universal intelligence. The intelligence that inspired geniuses like Leonardo da Vinci. This intelligence arrives fresh from the universe in every moment and we can all access it, as long as we keep the access routes clear.
 is everywhere now. It is too late to decide whether or not to use it. But we can determine how we use it and the quality we are in when we use it.){kind=link}