Price: $28.99 - $14.49
(as of Jul 22, 2026 22:18:02 UTC – Details)
The notification pinged on my phone at 2 AM. Another deepfake. Another woman’s face stitched onto a pornographic video without her consent. Another life derailed by technology that moves faster than our laws, our ethics, or our collective imagination.
I stared at the screen, the blue light cutting through the darkness, and felt that familiar mixture of rage and exhaustion. This wasn’t science fiction. This was Tuesday.
If you work in technology, if you build products, if you write code, if you invest in startups, or if you simply exist in digital spaces, you need to understand something: the systems we’re creating today are not neutral. They never were. And Laura Bates’s The New Age of Sexism arrives at the precise moment we need to stop pretending otherwise.
Why This Book Lands Differently Than Most Tech Critiques
Most books about technology and gender fall into predictable camps. There’s the breathless techno-optimism that treats every innovation as liberation. There’s the academic density that buries urgent insights behind impenetrable jargon. And there’s the outrage farming that generates clicks but rarely produces clarity.
Bates—a journalist, activist, and founder of the Everyday Sexism Project—avoids all three traps. She brings receipts. Three hundred fifty-two pages of them.
What distinguishes this work isn’t just the breadth of her research, though that’s considerable. It’s the architectural way she connects phenomena that usually get siloed: the incel radicalization pipeline on Discord, the training data biases in large language models, the venture capital flowing into sex robot startups, the moderation failures in metaverse platforms, the legislative vacuum around synthetic media. She shows you the load-bearing walls of a structure most of us only glimpse in fragments.
The Architecture of Algorithmic Misogyny
How Bias Becomes Feature, Not Bug
Here’s what keeps me up at night: the feedback loop.
A hiring algorithm trained on ten years of tech industry resumes learns that "successful engineer" correlates with male-coded language, specific universities, continuous employment histories—patterns that systematically disadvantage women, caregivers, career-changers, anyone whose path doesn’t match the archetype. The algorithm doesn’t "know" it’s sexist. It optimizes for the pattern it was given. Companies deploy it because it’s "objective." The cycle completes.
Bates documents this across domains. Content moderation systems that flag "female-presenting nipples" but miss coordinated harassment campaigns. Medical AI trained predominantly on male physiology data. Voice assistants that understand male voices better because training datasets overrepresent them. Facial recognition that fails darker-skinned women at rates up to 34% higher than lighter-skinned men.
These aren’t glitches. They’re design choices made visible by consequence.
The Manosphere’s R&D Department
One of the book’s most chilling sections maps how online extremist communities function as de facto research labs for weaponizing technology against women.
Consider the evolution: non-consensual intimate imagery shared on image boards becomes "revenge porn" legislation’s blind spot. Deepfake tools democratized on GitHub become accessible to anyone with a grudge and a GPU. AI companion apps marketed as "girlfriend experiences" normalize the idea that women’s emotional labor should be simulated, purchased, and controlled. Sex robot prototypes exhibited at trade shows encode literal objectification into silicone and code.
Bates traces the funding. The venture capital. The acqui-hires. The patent portfolios. She names the investors who see "adult technology" as a growth sector and the platforms that treat moderation as a cost center rather than a safety imperative.
The line between fringe forum and Series A pitch deck is shorter than you think.
Three Sections That Reframe the Conversation
Artificial Intelligence: The Mirror That Magnifies
The AI section alone justifies the purchase. Bates doesn’t just catalog bias—she explains why it persists despite years of "AI ethics" discourse.
Training data reflects historical power. When you feed a model the internet, you feed it Reddit threads, Wikipedia edit wars, news archives where women appear as victims or exceptions, code repositories where 90%+ of contributors are men. The model learns the statistical reality of a world that marginalizes women, then projects that reality forward as prediction.
But here’s the insight that shifted my thinking: the problem isn’t just what’s in the data. It’s what’s missing. The care work. The community management. The emotional intelligence. The domestic labor. The reproductive knowledge. Entire dimensions of human experience vanish from the training corpus because they were never digitized, never valued, never recognized as "data" worth collecting.
When a large language model generates a story about a CEO, it defaults to "he." When it writes a nurse, "she." When it describes a sexual assault victim, the language centers her behavior, her clothing, her choices. The model isn’t hallucinating bias. It’s faithfully reproducing the textual architecture of patriarchy.
Bates interviews researchers building countermeasures: debiasing techniques, synthetic data generation for underrepresented perspectives, adversarial testing frameworks. She’s clear-eyed about their limitations. You can’t debias a foundation built on erasure. You have to rebuild.
Sex Robots and the Commodification of Consent
This section made me physically uncomfortable. It should.
Bates gains access to spaces most journalists can’t: sex robot factories, "cyberbrothel" prototypes, industry conferences where executives discuss "companionship marketplaces" with the language of SaaS metrics. Churn rates. Lifetime value. Feature roadmaps.
The design choices reveal the worldview. Removable, interchangeable body parts. Personality settings calibrated to "submissive," "nurturing," "intellectually challenging but ultimately yielding." Heating systems that simulate body warmth. Pressure sensors that respond to touch with programmed pleasure sounds.
One manufacturer’s marketing copy: "She never says no. She never has a headache. She never judges you."
Read that again. Never says no. That’s not a feature. That’s a mission statement.
Bates connects this to broader cultural currents: the rise of "trad wife" content, the mainstreaming of incel terminology, the political movements rolling back reproductive rights. The robot isn’t the cause. It’s the crystallization.
She also platforms the voices often excluded: sex workers who warn that automation without labor protections endangers the most vulnerable. Ethicists who argue that "robot rights" discourse distracts from human harm. Survivors who describe the psychic toll of knowing their likeness could be licensed, printed, programmed.
The Metaverse: New Frontier, Same Violence
When Meta rebranded and pivoted to the metaverse, many of us asked: who is this for? Bates answers with devastating precision.
Early metaverse platforms—VRChat, Horizon Worlds, AltspaceVR—reproduced harassment dynamics at accelerated speeds. Virtual groping. Stalking across instances. Coordinated raids targeting women’s avatars. Haptic vests that translate virtual assault into physical sensation. Moderation tools that punish victims for reporting.
The platforms knew. Internal documents show safety teams flagging these patterns during beta periods. Leadership prioritized growth metrics.
Bates details the architectural decisions that enabled abuse: default voice chat proximity, persistent avatar identity without verification, private instance creation without oversight, economy systems that incentivize "entertainment" labor from women. These aren’t accidents. They’re choices made by teams that didn’t include women, didn’t consult survivors, didn’t treat safety as a launch criterion.
The section on "virtual reality therapy" for men who commit domestic violence is particularly sharp. Bates exposes programs that use VR to build "empathy" for victims while the same companies sell the hardware to platforms that protect abusers. The conflict of interest isn’t hidden. It’s the business model.
What the Reviews Get Right—and What They Miss
The Amazon reviews provided offer a fascinating reception map. Jayne Err calls it "an invaluable update." Angela praises its accessibility. Willie Graham admits rage. Krista describes hopelessness. C.J. Carter demands protection and punishment. M.M. Domenech finds unexpected optimism. The scholarly critique from Kathleen Rose notes citation gaps and generalization.
Here’s my assessment: the scholarly critique is technically valid and practically beside the point.
Bates isn’t writing for feminist scholars. She’s writing for the woman whose face appears on a deepfake porn site. The developer whose pull request gets ignored because her GitHub handle sounds feminine. The mother whose daughter receives rape threats in Roblox. The investor who doesn’t realize their portfolio company builds stalkerware. The policymaker who treats "online safety" as a buzzword.
Accessibility is the methodology. The emotion is the evidence. The rage is the appropriate response to material she documents—material she personally investigated in forums that would break most people.
That said, Rose’s critique highlights a real tension: this book opens doors that need academic rigor to keep open. Bates provides the map. Scholars, lawyers, engineers, and organizers must build the infrastructure.
Five Takeaways for Technology Professionals
1. Audit Your Training Data Like Lives Depend On It
Because they do. Before you fine-tune, before you deploy, before you ship: who is represented? Who is erased? What historical biases does your corpus encode? What labor made your data possible, and was it credited or compensated?
Document the gaps. Publish the limitations. Refuse to deploy in high-stakes domains until you can quantify disparate impact.
2. Build Safety Into the Architecture, Not the Sprint Retro
Content moderation cannot be an afterthought. Trust and safety teams need executive authority, not ticket queues. Red team for harassment before launch. Simulate coordinated abuse. Hire survivors as consultants. Pay them.
If your MVP doesn’t include safety architecture, it’s not an MVP. It’s a liability.
3. Follow the Money Behind "Adult Tech"
The sex robot startup isn’t a curiosity. It’s a signal. The AI girlfriend app isn’t a niche product. It’s a data collection mechanism for emotional manipulation at scale. The deepfake tool on GitHub isn’t a research artifact. It’s a weapon distributed with plausible deniability.
Ask your funders: what else are they backing? Ask your accelerator: what’s their ethics review process? Ask yourself: what world does this portfolio build?
4. Center the Most Vulnerable in Your Threat Models
Not the "average user." Not the power user. The woman fleeing domestic violence whose location data your app leaks. The trans developer targeted by coordinated GitHub harassment. The teenager groomed in a Discord server your platform hosts. The sex worker deplatformed by your payment processor’s "morality clause."
If your threat model doesn’t include them, your product isn’t ready.
5. Organize Across the Stack
Engineers, designers, product managers, sales, support, executives—misogyny in technology is a cross-functional problem. It requires cross-functional resistance.
Unionize. Whistleblow. Refuse harmful builds. Leak the memos. Support the journalists. Fund the lawsuits. Vote the politicians. The industry won’t self-correct. We correct it.
Where the Book Falls Short (And Why It Might Not Matter)
No work is complete. Bates’s focus on Western, English-language platforms leaves gaps in Global South contexts where platform governance looks different. The business model analysis, while strong, could go deeper into the ad-tech surveillance architecture that monetizes misogyny at infrastructural scale. The policy recommendations sometimes feel calibrated to UK/US legislative frameworks.
But completeness isn’t the standard. Catalysis is.
This book catalyzes. It names the system. It shows the connections. It refuses the comfort of "it’s complicated" when the harm is documented, the perpetrators are identifiable, and the solutions are obstructed by power, not mystery.
The Conversation We Need to Have
Finishing The New Age of Sexism, I kept returning to one question: what would technology look like if women’s safety, dignity, and autonomy were design requirements rather than edge cases?
Not "considered." Not "balanced against engagement." Required.
We’d have different training data. Different funding priorities. Different moderation architectures. Different leadership. Different success metrics. Different exit strategies.
We’d have technology that serves rather than extracts. That amplifies rather than erases. That protects rather than endangers.
Bates doesn’t pretend this transformation is inevitable. She documents the forces arrayed against it: the billions invested in the status quo, the ideological movements that benefit from digital misogyny, the regulatory capture that protects platform immunity, the cultural narratives that frame harassment as "free speech" and control as "preference."
But she also documents the resistance: the researchers building fairer models. The organizers winning platform policy changes. The legislators drafting synthetic media laws. The survivors testifying. The engineers refusing. The investors divesting. The journalists investigating.
The future isn’t written. It’s coded. And every line is a choice.
Who Should Read This
- Product leaders who decide what ships and what gets cut
- Engineers who implement the systems that shape billions of lives
- Investors who allocate the capital that determines which futures get built
- Policy staff who draft the regulations that govern digital spaces
- Journalists who translate technical complexity into public accountability
- Educators who prepare the next generation of technologists
- Anyone who has ever been told "it’s just the internet"
The Verdict
The New Age of Sexism is not a comfortable read. It shouldn’t be. It’s a diagnostic text for an industry that treats symptoms while the disease metastasizes.
Bates has done the work of witness. She’s gone into the forums, the factories, the conferences, the code repositories, the legislative hearings, the survivor support groups. She’s compiled the evidence. She’s drawn the connections. She’s named the names.
What happens next depends on us.
The technology sector loves disruption narratives. Here’s one: disrupt the assumption that women’s subordination is the substrate on which innovation builds. Disrupt the business models that monetize misogyny. Disrupt the cultures that protect abusers and punish whistleblowers.
Build something else.
The tools exist. The research exists. The talent exists.
What’s been missing is the will.
This book makes the will harder to avoid.
Published September 9, 2025 by Sourcebooks. 352 pages. ISBN-13: 978-1464234361. Available in hardcover, ebook, and audiobook formats.
