AI Bias: Who Holds The Power And How To Influence Change

By Rachel Lankester, Editor – with a lot of help from AI

AI is a passion of mine. I love to play around with it and see what answers it comes up with. I decided to challenge ChapGPT on its own biases. I asked: Is AI inherently sexist and ageist because it has been built using online resources that are inherently sexist and ageist?

I should have added in race but as a privileged white woman I didn’t immediately think of that – my own biases at play. After a bit of back and forth, Chat GPT asked if I’d like it to draft a blog about AI bias. I had also asked what I could do to help make AI a more representative, less biased tool. And who are the people with power who need influencing?

Here’s what it came up with. I reckon it’s pretty good. What do you think?

Is AI Biased — Or Are We? Why It Matters Who Builds the Future

We like to think of artificial intelligence as neutral — a kind of futuristic logic machine that tells us the truth, untainted by human emotion or prejudice. But that’s a dangerous myth.

The truth is, AI is trained on us. It learns from the data we create — from books, websites, job ads, medical records, and social media. And because our world is full of bias — including sexism, ageism, and racism — AI can learn and amplify those patterns unless we consciously intervene.

Algorithms Don’t Hate — But They Do Inherit

It’s easy to imagine AI as a villain in a sci-fi movie. But the real danger is more subtle: when AI quietly reflects and reinforces the inequalities that already exist.

It’s when:

• A hiring algorithm filters out older applicants because it “learned” that younger candidates were hired more often.
• A search engine mostly shows white men when you type in “CEO” or “scientist.”
• A healthcare tool underestimates the needs of Black patients because it was trained on a system that already underserves them.
• A voice assistant defaults to a female voice and subservient tone, reinforcing outdated gender roles.

None of this is intentional — and that’s precisely the problem. If we don’t notice it, we don’t fix it.

Who Gets To Shape AI?

AI isn’t just about technology — it’s about power. Who gets to decide what data is used? Who’s on the team designing the product? Who gets to say what “fairness” means?

Right now, the answers too often leave out women, older people, and racially marginalized groups — not just as users, but as creators, testers, and decision-makers.

What Can We Do?

This might sound overwhelming, but there are things we can all do — whether or not we’re engineers or policymakers.

• Ask questions: When you use an app or service powered by AI, ask how it was trained, and whether it was tested for bias.
• Speak up: If something feels wrong — an AI-generated image that stereotypes, a chatbot that assumes your gender, a hiring process that seems skewed — leave feedback. It matters.
• Support ethical design: Choose platforms and products that are open about how they approach fairness and inclusion.
• Push for change: Reach out to lawmakers, support organizations advocating for ethical tech, and stay informed.

A More Inclusive Future Is Still Possible

I believe AI can be a force for good — but only if we’re intentional about how we build it. That means recognizing that bias isn’t just a glitch in the system. It’s a mirror. And we all have a role to play in deciding what kind of world it reflects back to us.

Who Has Power — And How To Influence Them

If you’re interested in who has power and how to influence them, this is what ChatGPT told me:

1. Tech Companies and Product Teams

• Big players like Google, OpenAI, Meta, Microsoft, Amazon, and Apple shape how AI is trained and deployed.
• Influence tip: Feedback through their platforms matters. So do boycotts, user advocacy, and supporting smaller, more ethical alternatives.
• Employees also drive change — internal whistleblowers and employee groups have pushed companies to rethink harmful AI.

2. Governments and Regulators

• They shape the rules of the road for AI: what’s allowed, tested, or banned.
• Examples:
  • EU AI Act (emerging in Europe)
  • U.S. White House “Blueprint for an AI Bill of Rights”
  • UK AI Safety Institute
• Influence tip: Write to MPs or lawmakers, join advocacy orgs (like Access Now or Mozilla Foundation), or share calls for public feedback on proposed laws.

3. Researchers and Academic Institutions

• Universities help shape what’s considered “important” or “ethical” in AI — and what gets funded.
• Influence tip: Support inclusive academic work, amplify scholars who focus on marginalized voices in tech, and challenge elitist or narrow definitions of “AI progress.”

4. Investors and Venture Capital

• Investors push start-ups toward growth and profit — sometimes at the expense of ethics.
• But they can also influence more responsible design if they value long-term trust and social impact.
• Influence tip: Support or follow ethical investing groups (e.g., Omidyar Network, the Center for Humane Technology). Push for transparency in where money is going.

5. Media and Cultural Shapers

• How AI is portrayed — in news, movies, or ads — affects how people perceive it. Often, the problems of bias, fairness, or harm are totally ignored in favor of “futuristic” or scary narratives.
• Influence tip: Write op-eds, join community storytelling projects, or simply share and boost content that frames AI as a social issue, not just a tech one.

And if you’re concerned about this issue too, here’s ChatGPT’s thoughts on what we can do:

What You Can Do

1. Be a Conscious User and Critic

• Call out biased, stereotyped, or exclusionary outputs in the AI tools you use — even when it feels small.
• Leave feedback, especially when there’s an option to report harm, inaccuracy, or offensive content.
• Your voice helps build a pattern of concern that companies do pay attention to, especially if many users raise similar flags.

2. Support Ethical AI Projects and Organizations

• Use or promote tools made by groups committed to fairer AI (e.g., Hugging Face, AI Now Institute, Algorithmic Justice League).
• Follow, share, or fund the work of researchers or advocates doing this work. Some key voices are listed below.

3. Ask Questions Where You Spend Money or Time

• If a product, school, employer, or dating app uses AI, you can ask:
  • How do you test for bias?
  • Do you consider age, gender, and racial fairness in your design?
  • Do you have a diverse team?
• Even asking these questions shifts the culture — companies hate reputational risk and often start to care when users and consumers care.

4. Educate Others & Share Resources

• Host a conversation, write a blog post, or just casually bring up the topic — awareness creates momentum.
• Many people still think AI is neutral or “just math.” You can help shift that.

5. Advocate for Policy and Ethical Standards

• Contact your local representatives or join public comment processes when legislation on AI comes up.
• In the UK, EU, and US, policy is moving fast — and public input can guide how inclusive and fair those laws are.

People And Orgs To Follow

• Joy Buolamwini – founder of the Algorithmic Justice League; expert on facial recognition bias.
• Timnit Gebru – AI ethics researcher; co-founder of DAIR Institute (focused on marginalized perspectives in AI).
• Cathy O’Neil – author of Weapons of Math Destruction, about how algorithms reinforce inequality.
• Rumman Chowdhury – works at the intersection of AI ethics and policy; launched tools for bias audits.
• Hugging Face – a platform supporting open-source, inclusive AI.
• Mozilla Foundation – active in advocating for responsible AI and internet health.

So now I have an action plan! All done in less than 10 minutes. That’s the real power of AI I reckon…

Last Updated on August 4, 2026 by Editorial Staff