When an Algorithm Doesn’t See You

The first time I realized technology could overlook a person, it wasn’t dramatic. It was quiet—like a door that doesn’t slam, just refuses to open.

I was sitting in a cramped public library, the kind where the carpet smells faintly of dust and determination. Outside, rain tapped the windows in patient Morse code. I had just moved to the UK, carrying the careful optimism of an immigrant who believes good work will speak for itself. I uploaded my CV into an online hiring portal—one of those sleek systems that promises to match “top talent” with “high-impact roles.” The screen blinked, processed, and returned a polite rejection so fast it felt automated in more ways than one.

I told myself it was normal. I told myself I just needed more time.

The New Language of Opportunity

In STEM, we’re taught that data is neutral, that code is logical, that innovation is a straight line from problem to solution. But for people like me—navigating new professional norms, unfamiliar accents, and the subtle weight of being “other”—the path isn’t straight. It’s threaded with invisible checkpoints.

When I read about the All-Party Parliamentary Group (APPG) on Diversity & Inclusion in STEM launching a new project on AI equity, it felt like someone had finally turned on a light in a room many of us have been walking through by touch alone. The project aims to investigate barriers in Artificial Intelligence equity across STEM—barriers that affect who gets to participate, who gets access, and who benefits from AI-driven futures. It wasn’t just policy language; it was a mirror held up to experiences I hadn’t yet learned to name.

The Problem No One Calls a Bug

My complication wasn’t a lack of skill. I’d studied, built projects, learned new tools late at night. I could speak in Python and statistics with a fluency that surprised even me. Yet the gatekeeping felt algorithmic—like the system had been trained on a version of excellence that didn’t include my story.

In conversations with friends—data scientists, software developers, cybersecurity analysts, operations research students—I heard echoes: automated screening that filtered out “non-standard” career paths, recommendation systems that promoted familiar profiles, and the quiet discouragement of constantly proving legitimacy. AI wasn’t always the villain. Sometimes it was just the amplifier of old biases, scaled up and wrapped in a promise of objectivity.

What unsettled me most was the ethical shape of it: participation in AI isn’t only about who gets hired. It’s about who gets to build the tools that shape society—and whose values are embedded in them by default.

Small Acts of Resistance

So I started acting differently. Not grandly—just deliberately.

I sought communities that treated difference as data worth learning from, not noise to be filtered out. I joined meetups where my accent wasn’t a liability but a conversation starter. I volunteered for student projects that examined bias in training datasets and fairness in model evaluation. I learned to ask sharper questions: Who is missing in this dataset? Who benefits if the model is wrong? Who is harmed if it’s “mostly right”?

Reading about the APPG’s focus on AI equity gave those questions a public dimension. It suggested the issue wasn’t merely personal but structural—worthy of investigation, evidence, and change. That mattered. When institutions acknowledge a problem, it becomes discussable. When it’s discussable, it becomes solvable.

What I’m Taking With Me

The reflection that stays with me is simple: equity isn’t a soft add-on to innovation; it’s a condition for innovation to be trustworthy.

If AI is going to help allocate jobs, detect fraud, support healthcare decisions, or shape national security, then the people building it must include those who understand exclusion from the inside. Immigrant professionals don’t just bring technical skills—we bring lived experience of navigating systems that were not designed for us. That perspective is not a disadvantage. It’s a form of expertise.

A Door That Opens Both Ways

My resolution isn’t that every application suddenly worked or that bias disappeared. It’s that I stopped interpreting rejection as proof I didn’t belong. Instead, I began to see belonging as something we build—through policy, through practice, and through the courage to question “neutral” systems.

I’m applying to college because I want to study STEM in a way that keeps people in the frame. I want to help design AI that doesn’t just perform well on benchmarks, but behaves responsibly in the real world. The APPG’s project on AI equity feels like a signpost: the future is being negotiated right now, and I want to be part of the group that insists fairness is not optional.

Reference / Source Link

British Science Association. “APPG on Diversity & Inclusion in STEM launches new project on AI equity.” https://www.britishscienceassociation.org/news/appg-on-diversity-inclusion-in-stem-launches-new-project-on-ai-equity