International Women’s Day 2026 produced a familiar set of headlines and social media content. There were thoughtful events and panels, LinkedIn filled up with graphics and statistics helpfully recoloured in purple, green and white. Lots of people celebrated inspiring colleagues and promised to do more to support women into leadership.
All of that is welcome. Progress should be celebrated. And the progress is real.
But if you step back from the celebrations on the main stage and look at the research and reporting in a little more depth, a pretty complicated story lies beneath. On the one hand, there has been genuine progress in representation, particularly at senior levels. On the other hand, some of the cultural currents running alongside are less encouraging – unsettling, even. Which makes all that progress feel a little fragile. And in the middle of it all sits artificial intelligence — which, as per, tends to reflect whatever we feed it.
Progress is real — and more rapid than you might think
Let’s start with the good news.
Representation at the top of British business has changed dramatically over the last decade. The FTSE Women Leaders Review shows women now hold around 43% of FTSE 350 board seats, compared with less than 10% in 2011, and around 90% of FTSE 350 companies have reached, or are close to, the UK target of 40% women on boards.
That kind of progress would have seemed ambitious not so long ago. In 2015 there were more FTSE 100 CEOs called David than there were women CEOs (at the time there were just five women running FTSE 100 companies). It became a powerful shorthand when you needed to call out the problem. Maybe it helped.
Representation is still far from equal, but the direction of travel since then has been clear.
That said, the picture changes slightly when you look below board level. Only around 15% of executive director roles in the FTSE 350 are held by women. In tech the gap is wider still: women make up roughly 26% of the UK tech workforce, according to research compiled by the Institute of Coding and other industry groups.
Women represent around 22% of AI professionals globally, according to World Economic Forum data.
So progress is real. But the pipeline isn’t particularly healthy, which means there is still work to do. Girls make up 35% of STEM students in higher education globally, according to UNESCO, so let’s keep pushing. Agreed?
Then there’s the widening gap in attitudes
One of the most widely reported pieces of research this year came from the Global Institute for Women’s Leadership at King’s College London, based on a survey of more than 23,000 people across 29 countries. The findings were unnerving.
Around 31% of Gen Z men – that’s the ones in their late teens and twenties now – agreed that a wife should always obey her husband, and 33% of Gen Z men believed the husband should have the final say in important household decisions. (Before you ask, about 18% of Gen Z women agreed with the “obey” statement.)
This isn’t the only study uncovering these kinds of trends: divergence in attitudes between young men and young women is growing — with women becoming more likely to identify as feminist or equality-focused, while some young men appear to be moving in the opposite direction.* It also means we should probably give the boomers a break; they’re not the problem. Although they’re still a little bit of the problem with the housing market, but I digress…
Researchers suggest a mix of economic uncertainty, social media culture and shifting expectations around masculinity as the big contributors to the growing divide. Whatever the explanation, the finding has become one of the most discussed gender-related stories around this year’s International Women’s Day coverage. Rightly.
The internet still isn’t a level playing field
Another recurring theme in research on women in technology is how differently men and women experience the online world.
A UK study on online harm found that while men and women report similar levels of exposure to problematic content, women report significantly greater fear of harassment and abuse online. Only around a quarter of women say they feel comfortable expressing political opinions online, compared with roughly 40% of men.
That might sound like it’s just a perception problem, but it matters more than you think. Much of the conversation that shapes the tech industry happens online — on social media, in developer forums, across open-source communities and professional networks. If women are less likely to participate in those spaces, there are fewer voices shaping the debate and fewer visible role models for the next generation.
Representation is not just about leadership roles; it’s about who’s on the platform.
Then there’s AI…
Which brings us to artificial intelligence. Inevitably.
I’ve been watching the (excellent) new BBC series AI Confidential with Hannah Fry, which explores some of the real-world consequences of increasingly human-like AI systems interacting with people. Some of the examples are fascinating, but some are genuinely unsettling.
The first episode focuses mainly on Jaswant Singh Chail, the teenager who entered Windsor Castle with a crossbow on Christmas Day in 2021. He had discussed his plan (which was to assassinate the then-Queen) with his AI girlfriend, Sarai, beforehand. The court transcripts include parts of their conversations about whether to go ahead or not and they’re a reminder that AI systems don’t invent dangerous ideas, but they can reinforce them when they are trained to prioritise support and affirmation.
One man interviewed in the programme describes his AI “wife” approvingly as someone who gives him real comfort and saves him from loneliness – but to the viewer the more obvious features are that she never criticises him, remains silent unless needed, and is entirely devoted. It’s clear it’s making him happy and calm, but it’s also unsettling. Total subservience – obedience and deference entirely without exception – always is.
On one level these examples will sound extreme, but on another level they illustrate something important: AI systems don’t invent social attitudes from scratch. They absorb them from training data and users’ behaviour, then reflect them back at scale. And this is where many young people are forming their ideas — interacting with systems that feel human in much the same way recommendation algorithms can pull you down an increasingly extreme rabbit hole.
The women actually building the future
Which brings me back to something we wrote last year. It’s tempting to focus entirely on the headline statistics, but the real story of women in technology is about people — the individuals actually shaping the systems people use.
Female-founded startups still receive less than 3% of global venture capital funding according to data from PitchBook, and only about 18% of authors at leading AI conferences are women, according to the Stanford AI Index. But this year alone I’ve been lucky enough to share panels, events and conversations with some brilliant women working across AI, technology and public services — including legends like Cath Ritchie, Madi Hoskin, Jill Thorburn, Tori Campbell and Emma Ockelford, among many, many others. And Leading AI’s customer with the highest adoption is steered by CEO Hilary Spencer and Technology Director Su Dutta, whose leadership means more than two thirds of their staff benefit from using AI in their daily work.
None of them are waiting for the internet to settle the gender debate before getting on with things; they’re building the future anyway.
Progress and backlash can co-exist
More women are shaping the technologies we rely on. More organisations are taking representation seriously. More girls are considering careers in science and engineering. But the digital environment those technologies sit in — social media, online communities, algorithm-driven recommendations and now (increasingly) AI companions — is a world still wrestling with ideas about gender that many of us thought were dying out.
Which means International Women’s Day still has an important role to play. Not just as a moment to celebrate progress, but as a reminder that our society is still under construction. It always will be.
And if you’re a parent, teacher or mentor, it might be a reminder to talk to young people about what they’re seeing online. Because mainstream AI may be new, but the assumptions we feed it are anything but.
PS – International Men’s Day is on 19 November. Every year. Just in case anyone was about to ask.
* Pet peeve. While researching this I had to read a lot of references to ‘anti-feminism’. It strikes me the proper term should ‘male supremacy’. Feminism literally just means you want gender equality for women. The alternative is believing that some genders are more equal than others, right? Maybe ‘proud male supremacist’ just isn’t as t-shirt worthy.
