When You Are the Only One
- Sonia Brown MBE
- 1 day ago
- 10 min read
When You Are the “Only” One: The Hidden Barriers Facing Women in AI and STEM
AI is being built at speed, but opportunity is not moving at the same pace. Too many women are still expected to “break into” STEM while navigating systems that were never designed to recognise them as default talent. For Black and women of colour face in AI, the challenge is not one barrier, it is the stacking of barriers. Credibility pressure, visibility politics, unequal sponsorship and the quiet tax of always having to prove you belong.
Women in STEM are not underrepresented because they lack ambition or ability. They are underrepresented because the pipeline narrows at every point where power, sponsorship and progression decisions are made.
In the United Kingdom, women make up 27.6% of the core STEM workforce (over 1.4 million women), an all-time high, but still far from parity in the sectors shaping the future economy.
The AI picture is even more revealing. A UNESCO-backed synthesis of recent research highlights that women account for only around 18% of authors at leading AI conferences, alongside a higher-education landscape where more than 80% of AI professors are men.
That is not only a representation gap, it is a knowledge and influence gap, because publishing and professorship drive who gets funded, cited, hired and promoted.
Then there is the “everyday access” gap inside organisations. In McKinsey and LeanIn.org’s latest Women in the Workplace findings (published December 2025), only 21% of entry-level women say their managers encourage them to use AI tools, compared with 33% of men at the same level. When women are less encouraged to build AI fluency at the start, the leadership and innovation gap later is not mysterious, it is manufactured.
For Black and women of colour, these dynamics are often intensified by the additional burden of being both under-sponsored and hyper-scrutinised. In the United Kingdom, BCS reporting highlights how stark intersectional underrepresentation can be in technology roles, including the finding that less than 1% of IT specialists are Black females. That single data point helps explain why many BAME women in AI describe not just a career ladder, but a career maze.
Before we move into the specific barriers, it is important to name what often goes unspoken. The experiences of Black women and women of colour in STEM and AI cannot be understood through a single-lens gender narrative.
They sit at the intersection of race, gender, class, culture and power, where bias is not always loud, but deeply structural. These women are navigating industries that celebrate innovation while quietly reproducing exclusion, environments where merit is praised in theory but mediated in practice through access, sponsorship and proximity to decision-makers.
What follows is not a list of complaints, but a pattern of systemic friction points that consistently slow progress, drain talent and shape who is seen as credible, promotable and “future-ready” in the technologies defining our world.
Gender Bias and Discrimination
Studies show that gender bias persists in STEM fields, impacting hiring, promotions and recognition. Women often face scepticism regarding their capabilities and are underrepresented in leadership roles.
According to a report by the UNESCO Institute for Statistics, women account for only 30% of the world's researchers. In the AI field specifically, the 2020 Global AI Talent Report indicated women make up just 22% of AI professionals.
Black and women of colour face additional biases due to racial stereotypes, leading to fewer opportunities and less support in career advancement. This dual disadvantage impacts their presence and progression in STEM fields.
Lack of Role Models and Mentorship
The dearth of women in senior positions within STEM discourages young women from pursuing or persisting in these careers. Mentorship is crucial for career development but less accessible to Black and women of colour.
A study published in the Journal of Women and Minorities in Science and Engineering found that mentorship significantly increases the retention of women in STEM. However, only 26% of STEM graduates are women, highlighting the mentorship gap.
This has a massive impact on Black and women of colour often have less access to mentors who can relate to their unique experiences, which diminishes the guidance and encouragement they receive throughout their careers.
Workplace Culture and Inclusion
The workplace culture in STEM fields often lacks inclusivity, leading to feelings of isolation and alienation for women and BAME individuals, especially in predominantly male environments.
A Pew Research Centre study found that 50% of women in STEM jobs considered gender discrimination a major hurdle, compared to just 19% of men.
Discriminatory microaggressions and systemic racism contribute to a non-inclusive culture. Black and women of colour report higher rates of workplace discrimination and less job satisfaction compared to their white peers.
Pay Gap and Economic Inequality
There is a significant gender pay gap in STEM fields, with women earning less than their male counterparts. This gap is often wider for Black and women of colour due to intersecting biases.
The U.S. Department of Commerce reports that women in STEM earn 84 cents on the dollar compared to men. Black and women of colour earn even less, with Black women earning 62 cents and Hispanic women 54 cents, according to the National Partnership for Women & Families.
Economic inequality limits career advancement and financial independence, affecting long-term career opportunities and personal development for Black and women of colour in STEM.
Retention and Attrition
Many women leave STEM careers due to gendered experiences that reduce job satisfaction, such as lack of advancement, work-life balance issues and workplace discrimination.
A National Centre for Women & Information Technology report found that women leave tech roles at a 45% higher rate than men. This attrition is often due to inhospitable work environments.
Black and women of colour are particularly vulnerable to high attrition rates due to compounded stressors from racial and gender biases, leading to a significant loss of diversity in STEM fields.
Given these challenges, here are a few suggested next steps to address and support the involvement of women, especially Black and women of colour , in STEM and AI.
Increased Polarisation and Backlash
The weaponisation of Diversity, Equity and Inclusion (DEI) initiatives by influential figures like President Donald Trump and Elon Musk has had significant impacts on Black and women of colour across various sectors, including those in STEM and corporate environments. Here are some of the key impacts:
The politicisation of DEI efforts often leads to increased polarisation where initiatives meant to foster inclusion are viewed sceptically or as overly political. This can result in backlash against programmes designed to support Black and women of colour, creating environments where these initiatives are minimised or dismissed.
The backlash can foster a more hostile workplace atmosphere, creating additional barriers and challenges for Black and women of colour seeking to advance in their fields.
Reduction in DEI Programmes and Funding
When influential leaders undermine DEI, it can lead to cuts in funding and support for these programs, impacting their effectiveness and reach. Organisations may feel pressure to scale back efforts to avoid controversies or accusations of partisanship.
Undermining of Racial Equity Goals
Public statements or actions that dismiss DEI as unnecessary or divisive can undermine broader racial equity goals, diminishing the perceived importance of addressing systemic inequalities.
Without robust advocacy and commitment to DEI, the specific needs and challenges faced by Black and women of colour may not be adequately addressed, perpetuating existing disparities in representation and advancement.
Reinforcement of Stereotypes and Biases
Outspoken criticism of DEI initiatives can contribute to reinforcing negative stereotypes and biases, as dismissive attitudes towards diversity efforts can validate prejudiced perspectives.
This can exacerbate the microaggressions and biases Black and women of colour experience, further hindering their sense of belonging and ability to thrive in professional settings.
Influence on Corporate and Institutional Policies
The rhetoric and policies of high-profile individuals can influence broader societal attitudes, impacting how corporations and institutions prioritize DEI in their strategies and operations.
If businesses and educational institutions de-prioritise DEI, Black and women of colour may find fewer allies and advocates in leadership positions willing to champion diversity and equity causes, limiting their career advancement opportunities.
From Barriers to Blueprint: What Actually Works Now!
1. Mentorship That Transfers Power, Not Just Advice
Mentorship remains one of the most effective retention and progression tools, but only when it moves beyond informal guidance. For Black and women of colour in STEM and AI, mentorship must be structured, resourced and outcome-driven. An upgraded approach could be to:
Pair mentorship with sponsorship, where senior leaders actively advocate for visibility, stretch roles and promotion.
Build cross-organisational mentorship pools to reduce dependence on isolated allies within a single firm.
Measure success by career mobility outcomes, not participation numbers.
This reframes mentorship from emotional support to career infrastructure.
2. From “Diversity Training” to Decision-Making Accountability
In its traditional form, diversity training has lost credibility. One-off workshops focused on bias awareness and confirmation are easily dismissed, politically targeted and rarely change behaviour. The upgrade is not more training, it is decision accountability.
What works now:
Replace generic DEI training with role-specific equity intelligence for managers, hiring panels and senior leaders.
Embed inclusion into how decisions are made, not how values are discussed:
Who gets access to AI tools?
Who is nominated for high-visibility projects?
Who is promoted, paid and protected when mistakes occur?
Tie inclusive behaviours to performance reviews, leadership incentives and risk governance.
This approach shifts DEI from ideology to operational risk management and leadership competence.
3. Policy Advocacy That Is Quiet, Strategic and Embedded
In a politicised climate, public declarations alone are insufficient and often fragile. Policy advocacy must now be embedded into organisational systems, not positioned as optional values statements. The focus should be to:
Enforce pay transparency and audit mechanisms that automatically surface disparities.
Codify progression criteria so advancement does not depend on informal networks or “culture fit”.
Align equity goals with talent retention, innovation outcomes and regulatory compliance, particularly in AI governance.
This makes equity harder to reverse because it becomes structural, not symbolic.
Why This Upgrade Matters Now
In a climate where DEI is increasingly questioned, the most resilient equity strategies are those that:
Do not rely on moral persuasion alone.
Are defensible in business, governance and performance terms.
Protect talent without forcing individuals to carry the burden of change.
The future of inclusion in STEM and AI will not be decided by how loudly organisations speak about diversity, but by how deliberately they design systems that withstand political pressure, leadership turnover and cultural backlash.
Sustaining Progress in a Hostile Climate
Moving from diagnosis to durability requires more than programmes, it requires collective reinforcement. As external narratives attempt to weaken or politicise DEI, the blueprint must expand beyond internal fixes and into ecosystems of advocacy, protection and leadership development that can withstand pressure.
Advocacy and Awareness Building must now operate on two levels simultaneously. Internally, organisations need informed advocates who can translate equity into the language of risk, performance and innovation, ensuring that inclusion remains embedded in strategy even when public rhetoric shifts.
Externally, evidence-led storytelling and research dissemination help counter simplistic narratives by demonstrating that DEI is not ideological excess, but a structural response to measurable inequality. This dual approach reduces the vulnerability of inclusion efforts to political cycles and leadership changes.
Strengthening Support Networks is no longer optional, it is a stabilising mechanism. For Black women and women of colour navigating STEM and AI, networks function as shared intelligence systems, offering insight, protection and momentum in environments where informal access is uneven.
When these networks are recognised, resourced and connected to opportunity pipelines, they move from being informal coping spaces to becoming engines of retention, confidence and collective advancement.
Empowerment and Leadership Training completes the blueprint by shifting the focus from survival to influence. In an era where visibility carries risk as well as opportunity, leadership development must prepare Black and women of colour to operate at decision-making levels with authority, strategic clarity and resilience.
This means equipping them to lead complex teams, influence policy, shape ethical AI governance and advocate for equity from positions of power, rather than from the margins.
Together, these extensions reinforce the original blueprint by ensuring that progress is not dependent on goodwill alone. They recognise that inclusion in STEM and AI will only endure if it is defended through advocacy, sustained through networks and secured through leadership.
This is how barriers are not only removed, but replaced with structures that allow talent to enter, stay and lead, even when the climate becomes less hospitable.
Why This Moment Demands More Than Resilience
Taken together, these findings reveal a truth that can no longer be softened or side-lined. The underrepresentation and attrition of Black and women of colour in STEM and AI are not the result of individual shortcomings, but of systems that consistently extract value without offering protection, progression or parity.
Gender bias shapes who is believed. Racial bias shapes who is sponsored. Workplace culture determines who feels safe enough to stay. Pay inequity signals whose labour is truly valued. Attrition is not a mystery outcome, it is a predictable response to sustained exclusion.
What makes this moment especially fragile is that these long-standing barriers are now colliding with a broader cultural backlash against Diversity, Equity and Inclusion. The public weaponisation of DEI by influential figures such as Donald Trump and Elon Musk has not occurred in a vacuum.
Their rhetoric has helped legitimise scepticism, embolden resistance and give organisational cover to quietly retreat from equity commitments. When DEI is framed as political excess rather than structural necessity, the consequences land hardest on those who rely on these mechanisms not for advantage, but for fairness.
For Black and women of colour in STEM and AI, this creates a double bind. They are expected to be resilient in environments that remain inhospitable, while the very programmes designed to address inequity are questioned, defunded or diluted. The result is not only slower progress, but deeper isolation, increased psychological load and a narrowing of leadership pipelines at precisely the moment when technology demands broader perspectives, ethical scrutiny and cultural intelligence.
The path forward, therefore, cannot rely on resilience narratives alone. It requires deliberate architecture. Mentorship that is not accidental, leadership pathways that are not opaque, pay systems that are not negotiable by proximity to power and DEI strategies that are embedded, defended and measured rather than performative.
Advocacy, strong peer networks and leadership development tailored for Black women and women of colour are not optional extras, they are corrective tools for a system that continues to reproduce inequality at scale.
If AI is shaping the future of work, healthcare, education and governance, then who gets to build it matters profoundly. Inclusion is not about optics. It is about safeguarding innovation itself. Until Black women and women of colour can enter, progress and lead in STEM without carrying disproportionate risk, the industry will continue to fall short of both its ethical promise and its intellectual potential.
Let us keep the conversation going. Like this post to signal that these conversations matter. Comment with your perspective, your experience or the action you believe organisations must take next. Share it with a woman in STEM, a leader shaping AI strategy or a decision-maker who needs to understand that inclusion is not a side issue, it is a leadership responsibility.
If this conversation matters to you, there is more where this came from. We continue to explore the realities shaping diversity, equity and inclusion across leadership, technology, health, policy and culture, with depth, evidence and lived insight rather than headlines.
Join the discussion, engage with future articles and be part of a community that is committed to moving beyond performative inclusion towards meaningful change here.
Comments