
Part of The One-Person Renaissance: A 12-Month Blueprint to Becoming the Most Educated, Empowered Woman
I was scrolling on Social Media and everywhere I looked it was a new Claude feature here and a new AI Assistant there. And I sat there with my phone, my very human brain and that particular autistic mix of hyper focus and delayed emotional response. It felt like the floor shifted a few millimetres under my chair. Not enough to make anyone else look up. Enough that my nervous system noticed. It was not panic. It was not outrage. It was the quieter thought that scared me more. Things are changing and fast.
I work in HR. I have a law degree. I have spent years learning how to read people, how to write policy, how to live inside the gap between what organisations say they value and what they actually do. And here are people handing there most personal data and thinking to a machine.
And it made me realise: If I do not understand what this technology is, I will be making decisions about it with no real way to push back.
That was the day I stopped treating AI as someone else’s specialist topic and started treating it as part of my actual life.
January in the One Person Renaissance series was about economics and finance and that month showed me that the systems shaping our money are not neutral at all. February took that same lens and turned it toward technology. Specifically artificial intelligence. What it is. What it costs. Who it serves. What it means for women like us who are trying to build careers and lives on purpose instead of on autopilot.
This is not a post from a tech person. I am not a tech person. I am an autistic, burnt out, rebuilding person who decided that staying ignorant was no longer an option.
Why This Is Not Optional Anymore, Especially If You Are a Woman in a Professional Career
Here is the part that does not get said in polite meetings.
AI is already involved in your career. Not as a future scenario. As a present tense fact.
It screens your CV before a human ever scrolls. It scores your performance data. It flags patterns in how you communicate. It is used to predict who might leave, who is a high performer, who looks like a good bet. And the people building these systems are, statistically, not people who look like most of us or live lives that feel like ours.
We already know there is a pattern of AI systems absorbing the bias in their training data. Amazon scrapped an internal recruiting tool in 2018 after they discovered it was quietly pushing down CVs that included the word “women’s” — in things like women’s chess club or women’s leadership programme. The model learned from ten years of past hiring. Ten years of mostly hiring men. The machine copied the humans. The humans had a bias. The machine scaled it up and sped it up.¹
Understanding AI is not about becoming a programmer. I have zero interest in writing code. This is about basic fluency. Enough that when someone in a meeting says an algorithm made a decision, you know what to ask next. Enough to know when to say no.
For women in professional careers, that fluency is not an extra confidence skill. It is self protection.
What AI Actually Is, Translated Into Real Life
I spent a lot of February pulling jargon apart, because jargon is one of the neatest ways to make people feel stupid and shut them out of the conversation.
So here is how I now explain it when I am talking to friends who also do not feel like tech people.
The Short Version
When most people say AI right now, they are talking about systems trained on huge piles of data (often intelectual property of people who did not give liscences for it) that learn to spot patterns and then use those patterns to produce something. Text. Images. Recommendations. Scores.
The tools you probably use already fit this description. Autocomplete in your email. The recommendations on your streaming service. The chatbot that appears when you are trying to get a refund. They all do their own version of the same move. They look at past data, find patterns and guess what should come next.
The newer wave, the large language models like ChatGPT, are trained on staggering amounts of text from the internet. They learn the statistical relationships between words. When you ask a question, the model is not sitting there pondering like a tiny digital philosopher. It is predicting, in very fast detail, what a plausible response looks like given everything it has seen before.
That matters. It is not conscious. It does not understand your question the way a human does. It is extremely good at pattern recognition and at stringing words together. That is not the same thing as intelligence, no matter what the branding suggests.
The Longer Version
Ray Kurzweil’s The Singularity Is Nearer argues that we are moving toward a point where AI will surpass human intelligence in most areas of life and that this will happen faster than most people expect.
Reading him felt like sitting with a brilliant, relentlessly optimistic uncle. He has big timelines. Big confidence. Some of his earlier predictions have been uncomfortably accurate, which makes it harder to roll your eyes and move on.
Max Tegmark’s Life 3.0 felt different. Less like a prediction, more like someone saying, “Here are the questions we should probably ask before we press go on this thing.” Tegmark treats AI development as one of the most consequential things humans have ever done, and argues that what happens next depends heavily on choices we are making right now, not in some distant future we can leave to our grandchildren.
Neither book is light. I did not sit in the bath with a scented candle and casually absorb them. But both gave me ways to think about AI that went far beyond the usual “will it take my job” panic. Which, I learned fairly quickly, is the wrong question anyway.
The Critical Thinking Lens: AI Is Not Neutral
The most dangerous thing about AI is not that it might become smarter than us. It is that we might hand it authority before we have asked who built it, what it was trained on and whose interests it was designed to serve.
This is where most of my February energy went. Also the part that is almost totally missing from glossy AI conference talks and LinkedIn posts.
AI systems are built by humans, trained on data created by humans and rolled out to serve the interests of whoever is paying for them. None of that is neutral. All of that shapes what the system does and who it harms.
Nick Bostrom’s Superintelligence is dense and academic and at points I had to put the book down and go for a walk. Not because of science fiction scenarios, but because he is describing real structural risks that serious researchers are genuinely worried about.
His central obsession is alignment. How do you make sure a powerful AI system pursues goals that are genuinely good for humans, rather than goals that look correct in a narrow technical way but clash badly with our actual values. It is the difference between “technically correct” and “this ruins lives.”
Nicholas Carr’s The Shallows looks at something smaller and closer. Not future superintelligence. Just the internet we already have and what it is doing to our brains. He writes about how the constant flood of information and the way digital media is structured changes how we read, think and focus. His argument is that we are trading depth for speed. We skim instead of sit. We snack instead of cook. The cost of that trade is slow. You do not notice it on a Tuesday afternoon, but over years it adds up. And I do believe we see the first results already, one only has to look at teenagers today.
Reading Carr alongside Kurzweil and Tegmark created this odd tension in my head. Some people are looking at civilisation level change. Carr is looking at what happens in your nervous system every time you grab your phone instead of letting your mind wander for five minutes.
Both levels matter.
The Environmental and Economic Costs Nobody Mentions at Conferences
When I put my critical thinking hat on here, I found something that really bothered me.
Training a large AI model eats a huge amount of computing power, which means a huge amount of electricity. A 2023 study estimated that training a single large language model can create as much carbon dioxide as five cars over their whole lifetimes.² The data centres that keep these systems running use billions of litres of water for cooling. The hardware depends on rare earth minerals that have to be mined, often in places where local communities carry the environmental and human cost.
None of this means AI is automatically wrong. I do not think that. But it does mean the story of AI as a clean, neutral, purely progressive tool is incomplete. There are real costs. They do not land on everyone equally. And they almost never appear in the slide deck when someone is selling you efficiency.
Maybe, and this is something I also read about this month, AI will advance to a point where it can create nanobots (yes, a la Iron Man) and these bots can become every atom or substance needed. A rare mineral or a body cell. At that point we will probably mine less and possibly be able to reduce energy costs. But in what state will our environment be by then?
What I Actually Learned This Month, Including the Thing That Unsettled Me
I spent part of February listening to two episodes of Diary of a CEO that I would recommend if you want to think more seriously about AI without needing a computer science degree.
The Roman Yampolskiy episode is the one that lodged itself in my brain. He is an AI safety researcher and his position is not gentle. He does not think we can build a safe superintelligent AI. Not “it will be tricky.” Not “we need more regulation.” He thinks the problem might be impossible to solve and that most people pushing ahead with AI development are not acting as if that might be true.
And I get him. A superintelligence is unprecedented and able to think of scenarios we cannot even phatom. So how do we protect ourselves if it doesn’t stay benign?
As someone who likes solutions and frameworks and neat systems, I found that hard to sit with. I did it anyway.
The Mo Gawdat episode felt emotionally different. Gawdat, who used to be Chief Business Officer at Google X, talks about AI with a kind of tired urgency. He is not trying to stop it. He treats its progress as inevitable. His focus is on the values we pour into these systems and whether the people building them are thinking carefully enough about what they are actually creating.
The BigDeal episode about building a business with AI employees was the most concrete and grounded. No sci fi. Just small business owners matter of factly talking about using AI tools to handle operational work. What struck me was not fear. It was how normal they made it sound. This is not a thought experiment for them. It is a Tuesday. That made my own career questions feel less hypothetical too. This is already happening. The question is not if it touches your field. The question is how soon and in what shape.
The Thing That Genuinely Unsettled Me
Here is the part that got under my skin.
It was not the headlines about jobs being replaced. It was not Bostrom’s alignment problems. It was something smaller and closer.
I was reading about how some organisations now use AI in performance management. Tools that monitor productivity, track communication patterns and generate neat little assessments of individual performance.
And I thought about me.
I am autistic. My brain works differently. I have spent years masking and adapting and performing a version of myself that looks “professional enough” in offices that were not built with people like me in mind. So imagine an AI system trained on data from neurotypical high performers assessing my email response times, my wording, how often I speak in meetings and turning that into a score. No context for sensory overload, for burnout, for the careful pacing that keeps me functional. No measure of the actual thought processes brought into every decision.
That did not feel abstract. It felt personal. And it reminded me that the people already on the edge of existing systems are usually the ones most exposed when a new layer of automated judgement arrives.

The Tension: Tool That Frees You vs. Force That Replaces You
I refuse to be either a cheerleader or a catastrophist about AI. Both positions are lazy. The honest answer is that it depends, on who builds it, who governs it, who has access to it and whether the people most affected by it have any say in how it is deployed.
This is the question I kept circling all month, usually while making tea or staring out of train windows.
On one hand, AI already makes my life easier in very practical ways. I use it to sketch first drafts of documents that would otherwise take me a whole evening. I use it to summarise long reports when my brain is already fried. I use it to generate options when I am stuck in black and white thinking. I use it to summarise and plan projects, I use it to fasten up research.
It really has given me back some time. And as a burnout recoveree, time and energy are my most guarded resources.
On the other hand, I have watched AI being used to create deepfakes with malicious intentions, stolen intellectual property and accusations of using AI for writing anything in proper grammar.
So my conclusion was simple and not very glamorous. The technology is not the root problem. The governance is. And governance is a human problem, which means it is a space where people like us, educated, tired, paying attention, working in real organisations, are needed.
I am not anti AI. I am pro awkward questions about AI. Those are different positions and the way people talk about this topic often collapses them into a loud, useless binary.
Let’s return to a topic many recognise in the AI debate: Is AI coming for our joby.
The short answer. Yes
The long one: one at a time. At first it will and is already hitting entry level jobs and lower income fields. Anything that is easily repeatable. Those have been the first jobs to go anywhere. We are not dictating our letters anymore ever since everyone at work uses a PC, we need less people working in archieves if they’re digital.
But AI will go further and include more and more jobs.
This month I learned that humans will probably become like the nobility in history. Adminstrating themselves and living a cultural and social setting. No more 9-5, no more work.
But what will be our way towars that? Who will we loose and leave behind?
What will our society do, to ensure a stable life and income for everybody?
Those are the questions we already have to ask. But somehow we don’t.
How I Actually Studied This Month: My February Routine
I want to be specific about how I studied because “I read four books and listened to six podcasts” can sound either impressive or fake. Especially if you are already stretched thin.
I did not sit down and read four books from start to finish. I read what I needed and gave myself permission to be strategic.
Weeks one and two: I started with Life 3.0. Tegmark is the most readable of the four authors on my list, so I began there. Thirty minutes in the morning before work with my coffee. No highlighters. Just a plain notebook where I wrote down questions that came up as I read.
Week three: I moved to The Shallows in the evenings. Carr’s tone is more personal and the book is shorter. It felt like a good counterweight to the broad, sweeping thinking in Tegmark. That same week I listened to the Yampolskiy and Gawdat episodes of Diary of a CEO on my commute.
Week four: I did not force every page of Superintelligence. I used the index and the chapter summaries and went straight to the sections that connected with the questions I had already written down. I listened to the audiobook of The Singularity Is Nearer on my commute. I also listened to the BigDeal AI CEO episode and blocked out an afternoon to actually experiment with several AI tools. No theory. Just testing. Where did they genuinely help. Where did they spit out confident nonsense.
The honest truth is that when you study something this complex and emotionally loaded, the goal is not to finish the reading list. The goal is to walk away with better questions than the ones you started with.
That is what February gave me.
Since then I have been experimenting with AI agents and prompts to make my life easier and it also rekindled my love of learning, because in these changing times the ability to grow and learn will become a key factor.
By the Numbers
- Women hold only 26% of data and AI roles globally, according to the World Economic Forum’s Global Gender Gap Report 2023.³
- The McKinsey Global Institute estimates that up to 14 million workers in the US alone could need to change occupations by 2030 due to AI-driven automation, with women in administrative and service roles disproportionately affected.⁴
- A 2024 LinkedIn Workforce Report found that AI literacy is now among the top five skills employers are looking for across industries , including sectors traditionally dominated by women, such as healthcare, education and HR.⁵
- Despite this, women are significantly underrepresented in AI ethics and governance roles, the exact spaces where decisions about how these systems are built and deployed are being made.
Q&A: The Questions You Are Actually Asking
Q: I am not a tech person. Is it too late for me to understand AI well enough to use it or protect myself from it?
No. And I say that as someone who spent most of her adult life assuming technology was for other people. The barrier to understanding AI is not technical skill. It is vocabulary and confidence. You do not need to know how to build a model. You need to know what questions to ask when someone tells you a decision was made by one. Start with one good book, I would recommend Life 3.0 and one good podcast episode, the Mo Gawdat Diary of a CEO episode is a strong entry point. That is enough to start.
Q: Should I be worried about AI replacing my job?
Honestly? Some roles will change significantly. Some tasks within almost every role will be automated. But the research consistently shows that the jobs most at risk are those built around repetitive, predictable tasks, not the jobs that require judgment, relationship management, ethical reasoning or contextual understanding. Though I do believe they will follow. The goal is not to avoid AI. It is to understand it well enough to position yourself as someone who works with it intelligently, rather than someone who is simply replaced by it.
Q: How do I know if an AI tool is actually good or just well-marketed?
Ask three questions. First: what was it trained on and by whom? Second: what are its known failure modes or limitations? Third: who bears the cost when it gets something wrong? If the company selling you the tool cannot answer those questions clearly, that tells you something important. I also look at where servers are located, e.g. if they’re in Europe and follow the GDPR. But that’s a personal decision.
Q: I feel overwhelmed every time I try to read about AI. Where do I even start?
Start with the feeling, not the information. The overwhelm is real and it is not a sign that you are not smart enough. It is a sign that the topic is genuinely complex and that most of the public discourse around it is either hype or panic, neither of which is useful. Give yourself permission to go slowly. One book. One podcast. One conversation with someone who thinks critically about it. You do not need to understand everything. You need to understand enough to participate in the conversation.
Keep Going With The One-Person Renaissance
The full 12-month curriculum lives here: The One-Person Renaissance: A 12-Month Blueprint to Becoming the Most Educated, Empowered Woman.
January was economics and finance, understanding the systems that shape our money and our choices. February was this. Each month builds a different layer of the same foundation: a woman who understands the world she is living in well enough to move through it with intention and to grow her skills.
If you want to follow along in real time, subscribe to the newsletter below. I share what I am reading, what I am questioning and what I am actually applying – without the performance of having it all figured out.
Footnotes
- Dastin, J. (2018). Amazon scraps secret AI recruiting tool that showed bias against women. Reuters. https://www.reuters.com/article/us-amazon-com-jobs-automation-insight-idUSKCN1MK08G
- Strubell, E., Ganesh, A., and McCallum, A. (2019). Energy and Policy Considerations for Deep Learning in NLP. Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics. https://arxiv.org/abs/1906.02629
- World Economic Forum. (2023). Global Gender Gap Report 2023. https://www.weforum.org/reports/global-gender-gap-report-2023
- McKinsey Global Institute. (2023). The Future of Work After COVID-19. https://www.mckinsey.com/featured-insights/future-of-work/the-future-of-work-after-covid-19
- LinkedIn Economic Graph. (2024). LinkedIn’s Jobs on the Rise and Workforce Confidence Reports. https://economicgraph.linkedin.com



































