Digital Bypassing: The Myth of Neutral Technology
*We’re bringing this back from the archives with minor updates in 2026. If you’d rather listen to the original episode from 2020, head over to the podcast here. Part 2 coming soon with latest AI developments. Subscribe to the newsletter to stay informed.*
Is technology racist?
Most of us believe that technology is neutral— a fair and equal space for all to exist freely. But over the past few weeks, I’ve been exploring the topic of racism and how it’s embedded within technology itself. I have been studying the work of three women in particular:
Ruha Benjamin, author of Race After Technology
Safiya Noble, author of Algorithms of Oppression
Joy Buolamwini, a writer and a filmmaker known for her film Coded Bias
All are activists and advocates for Black women and for ethics in technology.
To begin, let’s actually talk about enlightenment (simplified).
The highest goal of spiritual practice is to “wake up”: dissolve illusion and see reality. This ‘illusion’ or Maya refers to the appearance that reality is our physical bodies, our thoughts, and even our physical world.
It’s through teachings, years of self-inquiry, and years of rigorous practice that we discover that the true nature of every being is actually the same: limitless, all-pervasive, pure consciousness. And this divinity abides in every person, regardless of their physical appearance or their family ancestry.
So this “realization” or enlightenment is the discovery of the true divine nature of your own consciousness.
When I first started on my spiritual journey, this idea of enlightenment was enthralling. I thought that, by learning this, I had everything figured out.
This notion of a place where all beings are equal, all beings are divine, is such a beautiful ideal that I strived to only feel that way. I wanted to rise above the emotions that I perceived to be negative (like anger or jealousy) because they seemed to not be in alignment with enlightenment.
I got sucked into the philosophy of good vibes only and I vowed to just be grateful and just be positive. I pushed away everything that didn’t fit into this narrative, and I relied on the thought that if I was on this journey towards enlightenment, then enlightenment would save me. Of course it would, right?
But then I did the arduous work: the years (and years and years) of deeply studying myself both through meditation and through therapy.
I realized what I was actually doing was spiritual bypassing.
Coined by John Welwood, spiritual bypassing is a tendency to use spiritual ideas and practices to sidestep or avoid facing unresolved emotional issues, psychological wounds, and unfinished developmental tasks.
We tend to cling to “absolute truths” to dismiss human needs, feelings, psychological problems, relational difficulties, and developmental deficits. He says that when we’re spiritually bypassing, we often use the goal of awakening to rationalize premature transcendence– trying to rise above the raw and the messy side of our human-ness before we have fully faced and made peace with it.
One of his favorite quotes is “idealism is an act of violence.” Trying to live up to an ideal instead of being authentically where you are can become a form of inner violence if it splits you in two and puts one side against the other.
Idealism is an act of violence.
When we use spiritual practices to “be good” or to ward off an underlying sense of deficiency or unworthiness, it turns into a crusade.
So he asks, how do we ripen so that we become naturally ready to let go of clinging to the self? Just as a ripe fruit naturally lets go of the branch and falls to the ground?
The practices that help us cultivate wisdom and compassion help with this ripening, but if we’re using our practice to avoid feeling our life, it will stunt the ripening process rather than support it.
So many of us do this not because we’re “bad,” but because it’s human nature to move towards pleasure and away from pain. As I contemplated this idea of spiritual bypassing, I couldn’t help but think about technology in the same way. What we’re seeing playing out is what I’d like to call digital bypassing:
Digital Bypassing©: a phenomenon like spiritual bypassing; the tendency to believe in the neutrality of technology, to avoid awareness of the inequities that exist within it, and subsequently evade responsibility when it produces biased outcomes.
– © Prerna Manchanda
Whether by idealism or by propaganda of the for-profit tech companies, we’ve come to believe that technology, including AI, is neutral and therefore less biased than you or me (a human with flaws, subconscious biases, etc.). It’s the illusion that AI is perfect because it is inhuman. In the same way that it’s damaging (psychologically per person and systemically as a society) to engage in spiritual bypassing, so too is it damaging to say the internet is colorblind, for instance.
Digital bypassing is to cling to the positivity and idealism of a magical new tool instead of facing the unpleasant truth that our technology is not neutral because we (humans) built it, and we are not neutral. Meanwhile, real humans feel the real detrimental effects of continued systemic injustice.
Ruha Benjamin talks about how we really want to believe that the internet is a revolution that sits above society– a blank-slate, tool-at-our-fingertips, neutral magical helper… Rather than sit with the admission that this technology is not above our society, it is a reflection of it.
It’s not unbiased, neutral, objective, or just a tool, no matter how much we hope it is. And people are being hurt in the process. This isn’t to say that there’s necessarily a bunch of evil programmers at these top companies that are purposefully creating racist or sexist programs. But systemic issues persist when we bypass reality. And the reality is that technology is a reflection of society, which has sexism, racism, classism… and that’s coded into the technology just as it’s coded into us until we do active work to combat it.
If technology like AI is a tool for progress and advancement… To whose benefit, and who holds the power to shape it?
We sprint forward down this path without pausing to try to answer this question. We turn a blind eye to the imbalances and harm done along the way. We digitally bypass.
If we pause our use and consumption and zoom out, here’s what we’d have to examine:
Who is creating the technology?
What are the world views that are getting embedded into this system?
Who benefits from the technology, directly or indirectly?
Who is hurt by the technology, directly or indirectly?
Who holds the power to continue shaping its use and impact?
John Perry Barlow wrote A Declaration of the Independence of Cyberspace, and he suggested that we are creating a world that all may enter without privilege or prejudice, accorded by race, economic power, military force, or station of birth. That we are creating a world where anyone may express his or her beliefs, no matter how singular, without fear of being coerced into silence or conformity.
Ruha Benjamin counters this by saying that the web is not only intangible, but it is a physical space too. It’s material and all of its qualities and our experiences with it are as real as any other aspect of life. Access to it is predicated on telecommunications companies and internet service providers. Its users live on earth in myriad human conditions that make them anything but immune to privilege and prejudice.
Human participation in the web is mediated by a host of social, political, and economic access points. So we really have to ask, what is it that is hidden? What is being bypassed by this idealistic vision of technology?
We live in a world now that is becoming more and more automated, particularly when it comes to decision-making. We rely on AI and computers to make many of the decisions for us. Artificial intelligence is basically asking, how do we model in machines what human users are interested in and going to do? It’s predictive based on human behavior, which is anything but unbiased. Ruha Benjamin suggests that the question we must answer first is this: Is there only one theory of the mind? And whose mind is it modeled on?
In the same moment that defenders of the technology claim it’s depersonalized, others in our society including the most vulnerable are actively experiencing the harm of it (whether we’re talking about abstract systemic issues like the perpetuation of racist microaggressions, or the very real and deadly effects on humans and environment of data centers).
Along the way, discrimination becomes displaced, and accountability along with it.
Well, it was the AI that did it. That was a bug. We’ll look into why that happened. The AI doesn’t create or program itself. We created it. We are responsible for the ‘mistakes’ it makes. We must start inquiring what is being programmed that might not actually be conscious.
Ruha Benjamin says that progressive-sounding values can create a veneer in which we stop questioning science and technology— Like generative AI stealing from artists but giving the tagline that now anyone can make art – hiding behind a veneer of progressive ‘accessibility’ messaging. Again, even if done without malice, the harm occurs anyway. And so we continue down the path of blindly following the ideals of enlightenment or positivity without doing the uncomfortable work, without deeply going through the layers of healing and nuance.
Safiya Noble says that the notions of this “universal human being” that is stripped of their unique self so that they can stand for every person, serve as the basis for a belief in an “ideal of unmarked humanity”: so non-racialized, non-gendered, and without class distinction, and that those who have become marked by race or gender or sexuality are deviations from the universal human. The false narrative is that personal liberties can be realized through technology, because of its ability to supposedly strip us of our specifics and make us equal. So while this sounds great, just like “all beings are one” sounds great, just because we have the ideal, it doesn’t mean that that actually plays out when the systems themselves are filled with bias. For example, striving for color blindness makes people of color, particularly Black people, not be seen.
There are many examples of how racism plays out in obvious and subtle ways across various types of technology. I want to spend some time sharing some to further contextualize digital bypassing.
Where is Digital Bypassing already perpetuating or emboldening inequity?
There was a beauty competition that was run only by AI. There were 44 winners, and all but six of them were white, and there was only one finalist that had visibly darker skin.
A former Apple employee who was working on speech recognition for Siri asked to add African-American English to speech recognition, and his boss said no, because Apple products are only for a ‘premium market.’
A woman Googled ‘unprofessional hairstyles at work’, and there were only images of Black women and Black hair. The ‘professional hairstyles at work’ results showed only images of white women.
Allison Bland, a media specialist, shared how Google Maps directed her to turn onto Malcolm 10 Boulevard instead of Malcolm X Blvd.
Safiya Noble has a famous example of googling ‘Black girls’ and getting pornographic results. This also was the case with Googling Latin girls and Asian girls. Women of color have been notoriously sexualized and fetishized, to the extent that you don’t even have to use the word sex or porn to get these images. This is very telling of the framework under which these codes were written.
After her book came out in 2018, Google addressed it, and now when you search Black girls, you’ll find images of Black Girls Code founder Kimberly Bryant and Me Too movement founder Tarana Burke, and various pictures of the Black Girls Rock! and Black Girls Run! organizations.
So, is it a glitch? A mistake? Or is the technology working precisely as designed? We must name and examine these occurrences instead of bypass them, because it’s in these seemingly harmless examples that lie the foundation or give permission for the larger and more detrimental inequities to continue. We see this in a lot of areas, but we can start again with Google.
But this doesn’t stop at digital, indirect, or abstract harm. It can have direct deadly consequences.
For instance, with COVID-19, hospitals used AI to determine who gets resources like ventilators. AI helped determine who would receive life-saving care based on who is most likely to survive – so healthier and fitter folks had higher chances of receiving that resource.
But, we exist in an inequitable society with for-profit healthcare, meaning health is a luxury. Access to preventive care is a luxury. Regular checkups, doctor advice, being taken seriously or respected by doctors is a privilege afforded to those systematically benefiting from the inequity. In this instance, the AI was anything but neutral, balanced, or fair.
When it comes to criminal justice, we can look at supposedly neutral risk assessment tools. Ruha Benjamin noted, how do we quantify or qualify risk? It’s not actually an objective science; a human must encode their own opinions about what is risky and why.
Factors being used for criminal risk assessment included if the neighborhood that someone is in has a higher crime rate, the history of unemployment, and family criminal records. But across the board for all of these factors, predominantly Black neighborhoods have higher rates. When we create these supposedly neutral risk assessment tools, we are ignoring how racism has already shaped someone’s life and their chances and their privileges (or lack thereof). So the tool “answers” that their criminal risk is inherent to them as an individual, bypassing how racism is already embedded in the tool, in societal context, and already shaped the individual’s context. This makes its way to the training and education of law enforcement, where there’s deadly impact on Black communities.
There was also the case of Henry Davis, who was charged with destruction of property for bleeding on police officer uniforms after these officers incorrectly identified him as having an outstanding warrant and then beat him into submission.
This was a mistake of a facial recognition software that generally cannot accurately recognize Black people (or women in general) correctly because the technology was trained on recognizing the faces of white men.
Joy Buolamwini talks about this idea of the Coded Gaze: algorithmic bias creating exclusionary experiences and discriminatory practices, priorities, preferences, and prejudices of those who have the power to shape technology.
She has done tremendous studies on facial recognition software when she herself found that she, as a Black woman, was unable to be recognized by this software. In the many studies that she’s done, the performance of the software has been perfect when testing white men and has been exceptionally wrong for testing Black men and Black women. And she says that, because our “data = our destiny,” the “pale male” (white men) data sets are destined to fail the rest of the world. If we’re not intentional about being inclusive, we reinforce these existing inequalities. She founded the Algorithmic Justice League, which helps to create “safe face” companies: companies that have their facial recognition software tested to be inclusive of all races and genders.
Once again going back to Google, she points out that Google Photos recognizes Black people as gorillas and Black women (even famous Black women like Michelle Obama and Oprah and Serena Williams) have all been misidentified by Google algorithms to be men. She found that Amazon’s facial recognition software had a 15% inaccuracy rate.
It’s not just law enforcement that uses this facial recognition software or the software given by private companies like Amazon. For example, college campuses use automated software to decide who can come on campus and who cannot by assessing threat. UCLA used this software. But when a digital rights nonprofit audited the software, they found 58% false matches that were predominantly people of color. So as Ruha Benjamin asked, what happens when the system flags you as an intruder? Campus police comes; LA police comes.
What does that do to you if you’re falsely identified as an intruder, especially if you’re a person of color? This happens all the time across many different institutions. These technical “mistakes” often target and impact the people already most vulnerable to bias.
There’s no accountability. There’s little recognition that so much about technology is powered by money– the technology that shapes our minds and shapes what we attend to, which then becomes our reality.
On social media for instance, with these echo-chamber-like algorithms, or search engine result pages… the people who have money and the people who are already privileged get their ads, articles, and photos in front of more eyes.
Renée Diresta said, “Free Speech is not the Same Thing as Free Reach”. We have this illusion that what’s most popular gets to the top of the search results first. But the first results are not necessarily a result of what is most clicked on or most popular or most credible. It’s really about how much money has gone into getting the first position.
Further, even if what’s seen the most is what’s most popular, that does not make it simple or ethical or neutral. The pornographic industry, so much of which is nonconsensual, has billions to spend to optimize their content, letting advertising drive content exposure and popularity. There are many layers of the system being skewed and flawed.
So what do we do?
Real enlightenment is a lifelong process that requires deep self-study and years of rigorous practice. We don’t become enlightened simply by reading about it. We don’t become anti-racist just because we follow Black voices on social media. We can’t wait for society to suddenly become equitable, and we certainly can’t simply hope and pray that technology will somehow be our saving grace.
Technology will always be a reflection of society. We have an opportunity to look at technology to more deeply understand ourselves and the biases that live within each and every one of us, and within the society at large. We need to think about building technology to be more fair and equitable on every level– from who gets seen to thinking about even E-waste that doesn’t exploit people of color around the globe.
Benjamin says that rather than trying to make quick hacks or fixes, we have to uproot these systems and start over. It’s the same concept as modern policing with roots in slave patrol— reform will never right the system, only abolition. From scratch, we must think about what it is that will actually create safety and regulation in our communities.
If we don’t address the root causes, technology layers on policing and makes these decisions (or lack thereof) invisible to the public eye. Accountability disperses. And it becomes “the way it is”. When we had a chance to correct that.
“Change is a matter of priority,” Joy Buolamwini said. We can create technology to be whatever we want it to be. We have the tools to do better and to be better.
Many of us are awake to these problems but complacent because we think of tech as god-like or inevitable. But that complacency gets us further from the ideals of freedom and justice for all. It’s up to US to decide how to proceed. How we question what we think human beings are capable of, and then engineer accordingly, as Joy Buolamwini put it.
We don’t need to create a shiny new system to make us better. We must become better, on purpose, together, as we continue to grow as individuals and as a society. We must question technology, see it as a mirror, understand ourselves and take our learnings from life experience and be able to apply it back to the tools that affect us. To make that technology better.
If the goal of our technological evolution is collective enlightenment, or being fair and equal, those are excellent goals, but we’ll never reach them while bypassing the reality of the now. And now, there is embedded violence that we have created in our own minds and hearts and built into society, consciously or not.
We cannot ripen and release it if we bypass the ripening– the naming, the acknowledging, the inquiry, the hard decisions, the uncomfortable truths, the work.
Real peace and real equality require deep work and deep acceptance of all parts of the human and societal experience– not displaced accountability, but actually integrating the parts that we haven’t gotten right and absorbing them so that we can learn from them and be better.
Practices like loving-kindness and meditation and compassion techniques are helpful to get us to empathize with others and, at least mentally, get into a headspace that each and every one of us deserves the opportunity for health, happiness, safety, and freedom. But we have to actually do the work to make it all possible. Not in another time, in another space. Right here on earth, right now.
—
Read more about self inquiry in a digital age on the blog, listen on the podcast, or work directly with Prerna here.