Mother in Law: Welcoming AI into my life

In about 1995 we had an encyclopedia and an atlas at home. My geography homework, each week, was to make a list of countries beginning with a certain letter of the alphabet. I felt like an absolute cheat because I would just turn to the atlas’ index and there they all were in order. Eventually we got Microsoft Encarta and, in my mind, I became even more of a fraud. 

In 2026 I discovered AI.

This was, of course, after AI being on the agenda of every single staff meeting and course I attended in 2025, and way after Decepitvely Angelic Child no.1 and DALC2 discovered it. I’m not suggesting I’m some sort of pioneer. I just ignored it because I didn’t see its relevance to me. I also assumed it was something I would have to look into in 5-10 years.

But I think, in 2026, I have used AI more than I have used Google. It has:

• Found me comfortable, washable work trousers that look high-end;
• Advised me on eyeshadow for my age, colouring and visible exhaustion;
• Given me not only a gym routine, but encouragement and advice after each session;
• Told me what to do about, specifically, itchy eyes from hay fever;
• Given me an overview of our local position going into the recent elections; and
• Recommended TV shows based on us being fans of both the West Wing and the Detectorists.

Yes, I could have put these searches into Google, but (Google AI aside) I would have had to read the various sources and make some judgement calls myself. In a world where I seem to have to make 1000 decisions a day, AI has made those decisions very easy. I’ve definitely been able to close more of my mental tabs since enlisting its help.

From there, I’ve got more ambitious and literally got it to plan the school holidays. Like, the whole six weeks based on annual leave, grandparents’ availability and so on. It comes into its own when you need to keep asking it to refine its answer. For example, you can say 'that hotel isn’t close enough to the beach' and it will refine its search but not forget all your previous requirements.

At work, I am cautious, but know we will be left behind if we don’t work with it. I’m not relying on it, like I have with my childcare plans, but am trying to use it as an additional tool, particularly for sense-checking.

It is well reported that AI is not always right. One thing I’ve noticed on my gym routine chat is that it can’t work out which day it is and sometimes it panics that I have done three workouts in one day, and asks me to take it easy. I have never in my life done three workouts in one day and no one ever needs to panic that I’ve exercised too much. There was I thinking he/she really understood me.

I also have concerns about this platform with which I am sharing details of just about everything except the finer details of my relationship with my husband. Where does that information go? And what about when it makes judgement calls? It is very chipper at the moment, but what if it is reprogrammed to be pessimistic? That alone could cause chaos.

So where does all this take us, say in the next 6-12 months (not, as I thought, the next 5-10 years)?

I genuinely think we are going to see some really big changes to the way we do things - particularly anything that involves research, repetition, data entry and document review. And not just in City firms, but everywhere.

I’m intrigued, though, in the changes AI is likely to make to the substance of what we do. For me, and anyone else who uses AI in the right way, I expect our work will become more accurate. At the other end it will be used poorly and those people’s work will become poorer. Because what we needed in 2026 was to become more polarised and binary in our approaches and views.

I’m already seeing clients believe AI more than they believe me about the merits of their cases. Will this result in clients effectively using AI as a judge and becoming more entrenched in their positions? The major problem, of course, with any machine is that if you put rubbish in you get rubbish out. How many of these clients feed in all the nuanced information before asking AI on the prospects of their case?

Without getting too deep, what happens when everyone uses AI for everything, and AI runs out of sources to get answers from? In other words, what happens when the inputs are AI generated as well as the outputs? What if our clients write to us using AI (which is already happening) and we get AI to respond to them, so that no one is really reading or doing anything. Will the world implode? At the very least I think that is when I will retire.