AI Obsolescence Anxiety: The Psychological Toll on Women in Tech

WOMEN IN TECH • July 29, 2026

The Pull Request That Wrote Itself

Codie is staring at a code review she didn’t write. The pull request is clean, well-documented, and functionally identical to the approach she would’ve taken herself, except she didn’t take it. An AI coding assistant generated it in eleven seconds, the same task that used to anchor a solid afternoon of her week, the kind of work she built her entire promotion case on three years ago.

She approves the pull request. It’s good code. That’s precisely the problem. Nothing about this moment is a crisis by any external measure: her team ships faster now, her manager is thrilled, her quarterly numbers look excellent. And underneath all of that, something in Codie has gone quiet and cold, a feeling she doesn’t have a name for yet, except that it feels uncomfortably close to grief.

Across the country, Kira sits in a product strategy meeting where a colleague casually mentions that an AI tool now handles the exploratory data analysis that used to be Kira’s specific, hard-won expertise, the skill that got her hired, the skill she spent four years in a PhD program earning. No one in the room seems to register this as significant. To Kira, it feels like watching her professional identity get quietly reassigned to a piece of software while everyone keeps talking about the plan.

Codie and Kira are composites, reflecting patterns I see often in my work with women in technical fields, not case studies of specific individuals. Their experience has a name, even though it rarely gets discussed directly in performance reviews or team meetings: AI obsolescence anxiety, and it’s a distinct, identifiable psychological pattern, not simply garden-variety career stress.

What makes this guide different from a general discussion of AI and the future of work is its focus. This isn’t about labor economics or which industries will shrink. It’s about the specific, felt psychological experience of a driven woman in a technical role watching her own particular expertise, the thing she built her professional identity around, become visibly automatable in real time, and what that does to her sense of self on a Tuesday afternoon in a code review or a strategy meeting.

It’s also worth being clear about what this guide doesn’t claim. It doesn’t claim that AI tools are uniformly bad for careers in technical fields, or that the right response is resistance to adoption. Many of the women I work with are, at the same time, genuinely excited by what these tools let them build faster and are also grieving a specific loss underneath that excitement. Both things are true at once, and a guide that only validated the excitement, or only validated the grief, would miss the actual lived texture of this experience.

What Is AI Obsolescence Anxiety?

AI obsolescence anxiety is a specific form of anticipatory fear that arises when a professional’s core technical skills are being visibly and rapidly replicated by AI systems, in a way that’s observable, ongoing, and directly tied to that person’s specific area of expertise. It’s not identical to general career anxiety or generic fear of change.

What makes this particular anxiety distinct is its object: it isn’t a vague sense that “the job market is changing.” It’s the specific, visible replication of a skill that a woman spent years building her professional identity around. That specificity is what makes it land so much harder than abstract disruption anxiety.

There’s also a timing element that intensifies the pattern. Earlier waves of workplace automation typically unfolded over years, giving people time to see the change coming and prepare. The current wave of AI-driven skill displacement in technical fields is unfolding in months, sometimes in a single product cycle. A tool that couldn’t reliably perform a given task in the spring can perform it fluently by the fall. That compressed timeline leaves little room for the kind of gradual identity adjustment that slower disruptions allow, which is part of why the anxiety it produces tends to feel so acute and so hard to reason your way out of.

Why This Threat Feels Existential, Not Just Professional

Terror management theory, developed by Jeff Greenberg, PhD, Sheldon Solomon, PhD, and their colleagues, describes how humans manage the underlying anxiety of mortality by investing heavily in symbolic systems, including career identity and mastery, that provide a sense of enduring significance. When a core professional competency, one of those symbolic anchors, is suddenly and visibly replaceable, the resulting anxiety often exceeds what the practical, financial stakes alone would explain.

Dan McAdams, PhD, whose research centers on narrative identity, has written about how people construct coherent life stories organized around defining competencies and turning points. For a woman who has spent a decade building a narrative identity around being the person who understands the data, or the person who can debug anything, an AI tool performing that exact task at speed doesn’t just threaten her job. It threatens the internal story she’s been telling about who she is.

This is where the psychological stakes exceed the practical ones. Losing a task is a logistics problem: redistribute the work, learn a new tool, adjust a workflow. Losing the narrative role that task played in your self-concept is an identity problem, and identity problems don’t resolve through logistics. They require actively rebuilding the story, not just reassigning the task list, which is a slower, more disorienting process than most workplaces have any real structure for supporting.

How It Shows Up in Women in Tech

Codie’s response has been to work harder and longer, taking on adjacent tasks she doesn’t need to take on, staying later than the work objectively requires, as though sheer effort can outrun a tool that doesn’t get tired. This is an understandable response and also, in my experience, a losing strategy: it treats the problem as a productivity gap rather than what it actually is, an identity disruption that productivity alone can’t resolve.

Kira’s response has taken a different shape: withdrawal. She’s stopped volunteering for the exploratory analysis work she used to love, not because she can’t still do it well, but because doing it now feels tinged with a grief she doesn’t want to feel in front of colleagues. She’s quieter in meetings. She’s started, quietly, looking at job postings in adjacent fields, less because she wants to leave and more because leaving feels like the only way to escape the daily reminder of what’s changing.

Neither response is a character flaw. Both are recognizable nervous system and identity-protection strategies responding to a genuinely disorienting shift, one that’s happening faster and more visibly in technical fields than in almost any other professional context right now.

A third pattern I see often sits between Codie’s overwork and Kira’s withdrawal: a kind of frantic credential-chasing, enrolling in course after course, certification after certification, in adjacent AI-related skills, not out of genuine curiosity but out of a fear-driven need to stay one step ahead of displacement. This pattern can look, from the outside, like admirable initiative. From the inside, it often feels like running on a treadmill that keeps speeding up, with no point at which enough learning finally produces the felt sense of safety it was supposed to deliver.

In My Clinical Experience

In my clinical experience, AI obsolescence anxiety is one of the fastest-moving psychological patterns I’ve seen emerge in the women I work with in technical roles, precisely because the pace of the underlying disruption keeps outrunning anyone’s ability to metabolize it. A woman might come into a session having processed one version of this fear, only to discover a new tool has shifted the ground again by the following week.

What I notice most often is that women in these roles are reluctant to name the fear directly, because naming it can feel like admitting weakness in an industry that prizes adaptability above almost everything else. Codie took three sessions before she was willing to say the word “grief” about a coding task. Kira described her withdrawal for weeks as simply being “less into it lately” before she could name what the shift actually was.

I want to be direct about something I see clinicians outside this specific professional context sometimes miss: this isn’t simply a case of generalized anxiety with a tech-industry flavor. The threat these women are naming is concrete and, in many cases, partially accurate. Part of the clinical work is helping a woman hold both truths at once: yes, something real is changing in her field, and also, her nervous system’s response to that change is exceeding what the actual situation requires her to feel in any given moment.

I also see a specific grief pattern that doesn’t map cleanly onto other kinds of career loss. Losing a job is a bounded event with a clear before and after. This is different: it’s an ongoing, incremental loss, a little more of the old skill set becoming irrelevant every few months, with no clear endpoint at which the grieving can be considered complete. Clinically, this calls for a different approach than acute grief work. It calls for building a woman’s capacity to tolerate ongoing, low-grade loss over an extended period, rather than resolving a single, discrete event.

One thing I’ve learned to watch for directly is somatic activation that shows up disconnected from its source. A woman will come in reporting insomnia, jaw tightness, or a persistent low hum of dread that she attributes to general stress, and it’s only once we trace the timeline carefully that the pattern lines up precisely with a specific product release or a specific meeting where her displacement became visible. Naming that connection explicitly, rather than treating the physical symptoms as free-floating stress, is often what allows the real work to begin.

Competence Identity and the Trap of Being Irreplaceable

Many women in technical fields arrived at their careers with an implicit contract: work hard enough, become skilled enough, and you’ll be secure. AI obsolescence anxiety breaks that contract in a way few other career disruptions do, because it doesn’t punish underperformance. It can arrive regardless of how skilled or hardworking a woman has been.

This is part of what makes the anxiety so disorienting. The usual playbook, work harder, get better, become more essential, doesn’t reliably apply when the disruption is structural rather than personal. Amy Wrzesniewski’s research on job crafting, the ways people actively reshape their relationship to their work to preserve meaning, offers a more useful frame here than the instinct to simply double down on the threatened skill.

The women I’ve seen navigate this most successfully aren’t the ones who out-worked the disruption. They’re the ones who found a way to build a professional identity that includes, but isn’t entirely organized around, being the fastest or most technically skilled person in the room. That shift is genuinely difficult for women who have spent their careers proving competence in male-dominated technical spaces, where the margin for being seen as replaceable has always felt thinner than it does for their male peers.

There’s a specific trap worth naming here: the belief that the solution to obsolescence anxiety is simply becoming irreplaceable again, faster, more indispensable, one step ahead of the next tool release. This belief is understandable and, in a field moving this quickly, largely unworkable as a long-term strategy. Chasing irreplaceability in a domain where the tools themselves are advancing month over month is a race that resets before it can be won, and building an entire sense of professional worth on winning it leaves a woman perpetually one product update away from the identity crisis returning.

Both/And: Real Disruption and a Distorted Threat Response

It would be a mistake to treat AI obsolescence anxiety as purely a distorted, irrational fear that needs correcting through reassurance. The disruption is real. Entire categories of technical tasks are genuinely changing in scope, pace, and who performs them. Dismissing that reality does women in this position a disservice.

It would be equally a mistake to treat every wave of anxiety as a proportionate, fully accurate read of the actual situation. The nervous system, once activated by a threat to competence identity, tends to generalize: one displaced task can trigger a global sense that the entire career is at risk, even when the actual professional picture is more nuanced and, often, more stable than the felt sense suggests.

Holding both truths at once is the work. Codie needed to acknowledge that her specific coding tasks were, in fact, changing in ways that required a real strategic response, learning to work alongside AI tools rather than in competition with them, while also recognizing that the flood of catastrophic thoughts about her entire career being over was her nervous system’s threat response overshooting the actual situation.

Kira’s both/and looked different: she needed to grieve the real loss of a specific kind of work she loved, the exploratory analysis that had defined her professional identity, while also recognizing that her withdrawal from meetings and colleagues was starting to cost her opportunities that had nothing to do with the disrupted skill itself.

• Name what’s actually changing in your specific role, with as much precision as you can, rather than letting the fear stay global and undefined.
• Grieve the real loss of tasks or skills that gave you meaning, rather than skipping straight to problem-solving mode.
• Build adjacent competencies deliberately, rather than either ignoring the shift or trying to out-work a tool that doesn’t tire.
• Separate your worth from your irreplaceability, a distinction that takes real, ongoing practice to internalize.

The Systemic Lens: Who Gets to Adapt, and Who Gets Replaced

AI obsolescence anxiety doesn’t land evenly. Women in technical fields already navigate a workplace culture that historically required them to prove competence more visibly and more repeatedly than male peers to be seen as equally capable. When a core skill becomes automatable, that same culture can be quicker to view a woman as replaceable and slower to invest in reskilling her for what comes next.

This isn’t paranoia. It reflects a documented pattern in how organizations allocate reskilling investment and stretch opportunities, patterns that already tend to favor employees, disproportionately men in technical fields, who are perceived as having more long-term strategic value. A woman whose specific skill is disrupted is, in many organizations, less likely to be offered the internal transition path that a male peer in an analogous position would be offered.

There’s also a compounding factor specific to women in tech: the same industry culture that made it harder to be taken seriously as technically credible in the first place is now the culture deciding who gets retrained and who gets quietly deprioritized. Taking the systemic lens means recognizing that individual resilience, while necessary, isn’t sufficient. Organizational and industry-level accountability for equitable reskilling matters too, and naming that clearly is part of processing this fear honestly rather than internalizing it as a purely personal failure to adapt fast enough.

Women who are also caregivers, or who are early in a family-building phase of life, face a further compounding pressure. Reskilling often requires discretionary time and cognitive bandwidth, evenings for a new certification, mental space to learn an adjacent domain, that isn’t distributed evenly by gender in most households. A woman navigating AI obsolescence anxiety while also carrying a disproportionate share of household and caregiving labor isn’t facing the same reskilling runway as a colleague without those demands, even when their job titles and skill disruptions look identical on paper.

None of this is a reason to abandon individual-level coping strategies. It’s a reason to hold them alongside a clear-eyed view of the uneven playing field they operate on, rather than treating a woman’s anxiety as purely a personal psychological quirk unconnected to the actual conditions she’s adapting within.

Organizations that want to retain their most skilled women through this transition have a real role to play here, one that goes beyond generic reskilling webinars. Transparent communication about which specific tasks are changing and why, equitable access to the internal projects that showcase adjacent skills, and managers trained to recognize identity-level disruption rather than dismissing it as an overreaction, all make a measurable difference in whether a skilled woman stays engaged through a disruptive transition or quietly starts looking elsewhere. The systemic lens isn’t only a diagnostic frame. It’s also a practical checklist for what actually needs to change at the organizational level.

How to Rebuild a Self That Isn’t Tied to Irreplaceability

The path through AI obsolescence anxiety isn’t becoming irreplaceable again, because in a field changing this quickly, that goal is neither realistic nor, in the long run, psychologically sustainable. The more durable path is building a professional identity that can tolerate ongoing change without collapsing every time a specific skill shifts in value.

This starts with naming the loss directly rather than skipping past it. Codie’s grief about the pull request that wrote itself was real and proportionate to what she lost: a specific, mastery-based relationship to a task that had anchored years of her professional confidence. Acknowledging that loss, rather than rushing past it toward the next skill to acquire, turned out to be a necessary first step rather than a delay.

From there, the work becomes building what Amy Wrzesniewski’s research calls job crafting: deliberately reshaping the parts of the role that still draw on distinctly human judgment, relational skill, and contextual understanding that current AI tools don’t replicate well, rather than either resisting the shift wholesale or trying to compete task-for-task with a tool built for speed.

Kira’s path forward involved a version of this: she moved from doing all of the exploratory analysis herself toward a role reviewing and directing AI-assisted analysis with an eye for the nuanced judgment calls the tool couldn’t make on its own, a real, valuable skill that drew on her expertise without asking her to compete directly with the tool on raw speed. It wasn’t the job she started in. It was a job that let her keep a version of the expertise she’d built without organizing her entire sense of worth around out-executing software.

Both Codie and Kira eventually arrived at a similar internal shift, even though their specific paths looked different. Each stopped measuring her professional worth primarily against the question can I still do this task faster than the tool, and started measuring it against a broader question: what do I understand about this problem, this team, this client, this system, that the tool has no access to. That reframe didn’t erase the anxiety entirely. It gave both women somewhere sturdier to stand while the ground kept shifting underneath the specific tasks.

None of this happens without some grief, and none of it happens instantly. But women who work through this deliberately, rather than either denying the disruption or letting it collapse their entire professional identity, tend to land somewhere sturdier than where they started: a sense of competence that includes adaptability itself as a core skill, rather than treating adaptability as evidence that the old competence didn’t matter.

It also helps to build in deliberate, structured reflection rather than letting the anxiety simply accumulate in the background of an already full week. A short, regular practice of naming what changed, what you actually lost, and what you’re building instead gives the nervous system a container for a disruption that otherwise tends to feel diffuse and unending. This doesn’t need to be elaborate. Ten minutes with a notebook at the end of a hard week, or a standing conversation with a therapist, coach, or trusted colleague who can track the pattern with you over time, is often enough to keep the fear from calcifying into a permanent sense of professional dread.

It’s also worth actively seeking out other women navigating the same shift, rather than processing it in isolation. Codie found real relief in a small, informal group of women engineers at her company who started meeting monthly, not to complain, but to name honestly what was changing and trade real strategies for adapting to it. The shared language mattered as much as any individual tactic: knowing she wasn’t the only one feeling this made the feeling itself less shameful and easier to work with directly.

If this pattern is showing up for you, working with a trauma-informed therapist or a trauma-informed executive coach who understands both identity psychology and the specific pressures of technical fields can help you build that sturdier footing more directly than trying to out-think the anxiety alone.

Warmly, Annie

Related Reading

• Greenberg, Jeff, Sheldon Solomon, and Tom Pyszczynski. “Terror Management Theory of Self-Esteem and Cultural Worldviews.” Advances in Experimental Social Psychology 24 (1991): 93, 159.
• Wrzesniewski, Amy, Justin M. Berg, and Jane E. Dutton. “Turn the Job You Have into the Job You Want.” Harvard Business Review 88, no. 6 (2010): 114, 117.
• McAdams, Dan P. “The Psychology of Life Stories.” Review of General Psychology 5, no. 2 (2001): 100, 122.
• van der Kolk, Bessel. The Body Keeps the Score. New York: Viking, 2014.

For more on related patterns, see Annie’s Women in Tech Resource Hub, and her guides to what financial trauma actually is, the IPO aftermath, achievement addiction after emotional neglect, and nervous system dysregulation in women who look fine.

About the Author

Annie Wright, LMFT

LMFT · Relational Trauma Specialist · Author, W.W. Norton 2027

Helping driven women finally feel as good as their résumé looks.

Annie Wright is an EMDR-certified licensed psychotherapist and relational trauma specialist with over 15,000 clinical hours, and she's been in practice since 2013. Trained in EMDR, psychodynamic, and somatic modalities, she is licensed in 14 U.S. jurisdictions and registered to provide telehealth in Florida (California, Colorado (telehealth only), Connecticut, the District of Columbia, Florida, Illinois, Maine, Maryland, New Hampshire, New Jersey, New York, Texas, Utah, Virginia, and Washington). Annie works with driven and ambitious women from relational trauma backgrounds, and everything she writes about is field-tested across thousands of clinical sessions. She is the founder and former CEO of Evergreen Counseling, a multimillion-dollar trauma-informed therapy center she built, scaled, and successfully exited, and is currently writing her first book, The Everything Years: Navigating the Pressure and Promise of Your Thirties, with W.W. Norton (2027). A regular contributor to Psychology Today, her expert commentary has appeared in USA Today, Forbes, Business Insider, Inc., NBC, and The Information.

Work With Annie

Medical Disclaimer