Reclaiming Dignity

Your phone buzzes.

A bank message. A job offer. A QR code on a notice. A cheap product ad. A voice that sounds like someone you know. A video with a familiar face. A family member asking, “Is this real?”

And now you pause.

You do not just live your day anymore. You verify it.

That pause is the story. Not panic. Not paranoia. Not some tired speech about people being too online. The everyday has become a security job, and ordinary people are paying for it in time, trust, and calm.

This is not about saying every new tool is dangerous. That would be too easy, and it would also be wrong. The sharper problem is this: trusted surfaces have become attack surfaces. The bank message, the social media ad, the QR code, the job offer, the voice note, the family emergency, the dating profile, the public figure in a video — all of it can still be real, but now you have to treat it as something that needs checking.

That checking is work.

The FBI’s 2025 Internet Crime Report found nearly $21 billion in cyber-enabled losses reported by Americans, with more than one million complaints. The report also included artificial intelligence-related complaints: more than 22,000 reports and nearly $893 million in reported losses. The FBI named fake social profiles, voice clones, fake identification documents, and believable videos of public figures or loved ones as tools used by scammers.

That does not stay inside a report. It lands in the hallway at 9:47 p.m., when someone still has their jacket on and has to decide whether to click, call, ignore, report, or ask another person for help.

UK Finance reported that criminals stole £1.28 billion through payment fraud in 2025. Authorised push payment fraud — the kind where the victim is manipulated into sending the money themselves — rose to £576.4 million. Most APP fraud cases still started online or through telecoms. That detail matters because it shows the shape of the risk: the criminal does not always need to break the door down. Sometimes they just convince you to open it.

That is why people behave differently now. They call the bank back through a number they find themselves. They stop before scanning QR codes. They do not trust a recruiter who will not use a company email. They screenshot messages before deleting them. They ask a sibling, partner, colleague, or grown child, “Does this look fake to you?” They search the company name plus “scam.” They check whether the webshop has a real address. They avoid moving conversations about money to WhatsApp or Telegram. They hesitate before buying from a sponsored ad, even when the product looks normal.

This is the shared reality. Everybody is being trained to become a small private fraud department.

Where men often feel it is in places where quick action is expected: payments, jobs, side-income offers, marketplace deals, investment opportunities, family tech help, and workplace approvals. A man selling something online now checks whether the buyer is real before handing over the item. A father gets a message from someone pretending to be a child in trouble and has to slow down the instinct to fix it immediately. A man looking for better work has to treat a flexible remote job offer as a possible trap before he can treat it as hope. A finance worker receiving a payment request from “the boss” has to ask one more awkward question before sending money.

That is not an offerrolle. It is behaviour. More pausing. More checking. More refusing urgency. More quiet shame if he gets it wrong, because the public story still says a competent man should not be fooled.

Where women often feel it is in places where image, access, and personal safety meet: social profiles, intimate images, dating, harassment, impersonation, and reputation damage. The TAKE IT DOWN Act requires covered platforms to provide a way to request removal of non-consensual intimate images, including deepfakes, and remove them and known identical copies within 48 hours after a valid request. That law exists because the risk has become normal enough to need a formal removal path.

Again, behaviour changes before the debate catches up. Women lock old photos. They remove public details. They think twice before posting pictures from a night out, a gym, a beach, a bedroom mirror, or a workplace. They screenshot threats. They ask friends to report fake accounts. They check whether a profile is real before replying. They share less, not because they have nothing to say, but because sharing now carries a different cost.

That is not moral superiority. It is risk management.

The sharp contrast is this: men are often pushed to verify before acting; women are often pushed to protect before being exposed. Both are living inside the same trust collapse, but the pressure lands differently. For men, the danger often hides inside responsibility, money, action, and being the person who is supposed to fix things. For women, the danger often hides inside visibility, image, access, and being made searchable or reproducible without consent.

The difference matters because a one-size-fits-all safety speech misses the actual behaviour. “Be careful online” is too thin. People are not just “online.” They are applying for jobs, paying bills, dating, buying food, helping family, running small businesses, raising children, and trying to get through a tired day without becoming someone else’s opening.

The uncomfortable truth is that awareness is not enough.

There is a limit to how much responsibility you can place on ordinary people before “be careful” becomes a polite form of abandonment. If scams are industrial, AI-assisted, platform-distributed, psychologically targeted, and designed to exploit urgency, loneliness, hope, duty, shame, and financial pressure, then the average person cannot be the only line of defence.

That does not mean people have no responsibility. It means the responsibility has been pushed too far downward.

The Federal Trade Commission reported that, in 2025, nearly 30% of people who reported losing money to scams said the scam started on social media, with reported losses reaching $2.1 billion — about eight times the 2020 level. Shopping scams were the most reported social media scams, often starting with ads. Investment scams caused the largest losses, and romance scams also heavily started on social platforms.

That changes ordinary behaviour. People stop trusting “sponsored” as a neutral label. They check comments. They search the company elsewhere. They wonder whether the discount is a trap. They hold back from the cheap thing they actually need because the risk of being tricked has become part of the price.

This is where the nice story breaks. Social media is sold as connection, discovery, opportunity, community, visibility, and access. In practice, the same machinery can help scammers target people by age, interest, shopping habit, loneliness, financial hope, and personal information. The platform feels casual. The manipulation is professional.

There is a positive truth here too, and it matters because otherwise the whole thing becomes doom. Concrete barriers work better than moral scolding.

Danish police reported that fake webshop cases rose sharply in 2025, while some other categories fell. The report also points to technical measures, including SMS filtering, as one reason some forms of fraud dropped. That is the important lesson: when systems build actual friction against criminals, ordinary people are less exposed. When systems only tell people to “stay alert,” the work is still sitting on the tired person with the phone in their hand.

Consumer Reports found in 2025 that several AI voice cloning products lacked meaningful safeguards against misuse, and that some tools could create voice clones from short or publicly available audio with weak consent checks. That does not mean every voice tool is bad. Some uses are legitimate: accessibility, editing, narration, dubbing. But the weak point is obvious. If a tool can copy a voice without real consent, then a family call, a business request, or a public figure endorsement becomes less trustworthy.

The worst damage is not only that fake voices exist. It is that real voices become harder to trust.

That is what fraud does when it scales. It does not only steal money from victims. It steals ease from everybody else. The person who has never been scammed still pays with hesitation. The person who has never had an image abused still pays by posting less. The person who has never sent money to a fake account still pays by slowing down every request. The family that has never been hit still makes a code word. The workplace that has never lost money still adds one more approval step.

And then people are judged for being slower.

That is the part we need to name. If ordinary people answer later, ask more questions, refuse urgency, lock accounts, avoid links, and distrust perfect-looking opportunities, it is not always cynicism. Sometimes it is adaptation. Sometimes low trust is not a personality flaw. Sometimes it is information.

Pew Research warned in 2026 that AI can also distort public opinion research itself, including through bogus respondents and AI-generated estimates that stereotype groups or flatten disagreement. That matters for any reality-facing project, because not every statistic is reality. Some numbers can become narrative with a spreadsheet face. The more automated the information environment becomes, the more important it is to ask whether a claim explains behaviour — or just justifies a story somebody already wanted to tell.

The public story still loves speed. Fast checkout. Fast approval. Fast reply. Fast onboarding. Fast payments. Fast content. Fast trust.

Private behaviour is moving the other way.

People are slowing down because they have learned that speed is where manipulation enters. They save receipts, check domains, ask who sent the message, look for original sources, refuse to be rushed, and choose known channels over convenient ones. The future may be more digital, but ordinary dignity now depends on the right to pause.

That pause is not weakness. It is a boundary.

The direction is not paranoia. A life where nobody trusts anybody is not dignity either. The direction is better shared responsibility: stronger platform accountability, better technical barriers, clearer reporting paths, real removal processes, safer defaults, and less shame placed on the person who was targeted by a professional system.

For the reader, the small everyday shield is simple enough to remember when tired: do not let urgency choose the channel. Use the channel you already trust. Call back through a number you find yourself. Ask for the next concrete step, the responsible sender, and the deadline. Save the screenshot before the message disappears. If money, identity, intimacy, or access is involved, slow it down.

The everyday has become a security job. That is the damage.

The answer is not to become cold. It is to stop letting systems call transferred risk “awareness.”

Who has moved the security work onto me here — and what small boundary can I set before I click, pay, reply, share, or believe?

---

Sources

FBI — Cryptocurrency and AI Scams Bilk Americans of Billions
Link: https://www.fbi.gov/news/press-releases/cryptocurrency-and-ai-scams-bilk-americans-of-billions
Summary: The FBI’s 2025 Internet Crime Report found nearly $21 billion in cyber-enabled losses, more than one million complaints, and nearly $893 million in AI-related reported losses. It also names voice clones, fake profiles, fake documents, and believable videos as scam tools.
Everyday impact: Normal messages, voices, and faces now require verification before trust.

UK Finance — Fraud remains a national security threat as criminals steal almost £1.3 billion
Link: https://www.ukfinance.org.uk/news-and-insight/press-release/fraud-report-2026-press-release
Summary: UK Finance reports £1.28 billion stolen through payment fraud in 2025. APP fraud rose to £576.4 million, with many cases starting online or through telecoms.
Everyday impact: People are manipulated into sending money themselves, which adds shame and makes urgent payment requests feel suspicious.

FTC — Reported losses to scams on social media eight times higher than in 2020
Link: https://www.ftc.gov/news-events/data-visualizations/data-spotlight/2026/04/reported-losses-scams-social-media-eight-times-higher-2020
Summary: In 2025, nearly 30% of people who reported losing money to scams said it started on social media, with $2.1 billion in reported losses.
Everyday impact: Social media stops being just a place to browse. Ads, profiles, and recommendations become things people have to verify.

Reuters / BEUC — Google, Meta, TikTok hit by EU consumer complaints about handling of financial scams
Link: https://www.reuters.com/legal/litigation/google-meta-tiktok-face-eu-consumer-complaints-about-handling-financial-scams-2026-05-21/
Summary: European consumer groups filed complaints under the Digital Services Act, saying major platforms failed to remove enough scam ads. Platforms defended their safety work, while consumer groups said only 27% of reported suspicious ads were removed in their test.
Everyday impact: Even when platforms remove huge amounts of scam content, ordinary users still face the scams that slip through.

FTC — TAKE IT DOWN Act enforcement
Link: https://consumer.ftc.gov/consumer-alerts/2026/05/what-will-ftcs-enforcement-take-it-down-act-mean-you
Summary: Covered platforms must provide a way to request removal of non-consensual intimate images, including deepfakes, and remove them and known identical copies within 48 hours of a valid request.
Everyday impact: Especially for women and young people, image-sharing becomes a safety decision, not just social expression.

Consumer Reports — Assessment of AI Voice Cloning Products
Link: https://www.consumerreports.org/media-room/press-releases/2025/03/consumer-reports-assessment-of-ai-voice-cloning-products/
Summary: Consumer Reports found that several voice cloning products lacked meaningful safeguards against fraud or misuse.
Everyday impact: Families and workplaces can no longer treat a familiar voice as automatic proof. Call-backs, code words, and written confirmation become normal.

Danish Police / NCIK — Årsrapport om it-relateret økonomisk kriminalitet anmeldt i 2025
Link: https://politi.dk/-/media/mediefiler/landsdaekkende-dokumenter/aktuelt/statistik-og-udgivelser/it-relateret-oekonomisk-kriminalitet/aarsrapport-om-itrelateret-oekonomisk-kriminalitet-anmeldt-i-2025.pdf
Summary: Danish police reported a sharp rise in fake webshop cases in 2025, while some other categories fell. The report also connects some reductions to stronger technical measures such as SMS filtering and safer systems.
Everyday impact: It shows that real protective infrastructure matters more than just telling people to “be careful.”

Pew Research Center — Do AI and bogus respondents threaten polling’s future?
Link: https://www.pewresearch.org/short-reads/2026/05/12/qa-do-ai-and-bogus-respondents-threaten-pollings-future/
Summary: Pew warns that AI-generated estimates and bogus respondents can distort public opinion research, stereotype groups, and flatten disagreement.
Everyday impact: Weak statistics can become public narratives, and ordinary people can be judged by numbers that do not actually describe their lives.