Reclaiming Dignity

You open your laptop after work and there it is again.

Another AI-powered tool. Another workflow. Another promise that this will save time. Another quiet expectation that you will learn it, test it, trust it, correct it, and somehow still deliver faster by Monday.

That is how the future arrives for a lot of ordinary people.

Not as freedom. As homework.

AI is being sold as the future. But ordinary people are increasingly being made into the adaptation layer.

That does not mean AI is fake. It does not mean AI is only bad. It means the real question is no longer whether the technology works. The question is who has to absorb the friction while everyone else calls it progress.

The shared reality is this: AI has left the screen.

It is now inside the power grid, inside job descriptions, inside local planning meetings, inside water systems, inside junior hiring, inside workplace expectations, and inside the quiet self-assessment people do when they ask themselves, “Am I still useful?”

The International Energy Agency expects data-centre electricity demand to roughly double by 2030, reaching around 945–950 TWh.

That sounds abstract until you remember that electricity is not an app. It has to be generated, transmitted, cooled, paid for and locally accepted.

Denmark’s Energinet has already announced a temporary pause on new grid connections while around 60 GW of new electricity consumption projects sit in the queue. In the EU, officials are looking at minimum energy-efficiency standards for data centres because projected capacity growth is no longer a niche issue. In the United States, energy regulators are asking how large new power users, including data centres, can connect without weakening reliability or pushing extra costs onto normal households.

This is what “the cloud” looks like when it comes back down to earth.

A person in a normal home does not experience that as an energy-policy debate. They experience it as a higher bill, a grid under pressure, a local construction project, a planning meeting nobody has time to attend, or a feeling that decisions have already been made before ordinary people are invited to react.

The cultural trick is that AI still gets spoken about as if it floats above everyday life. Innovation. Competitiveness. Upskilling. Green growth. Efficiency. These words are not always lies. Some describe real benefits. But they can also soften harder facts: more power demand, more water questions, fewer entry-level openings, more invisible training time, more private anxiety, more local disruption.

And when ordinary people question that, they are too easily treated as slow, scared or anti-progress.

That is lazy.

A lot of people are not rejecting the future. They are asking why the future keeps arriving as unpaid homework.

Men often feel this pressure through usefulness, status and replaceability.

Not as a clean victim story. More like a private behavioural shift. A man at work hears that AI will not replace him, but someone who uses AI might. So he starts using it after hours. He watches tutorials. He tests prompts. He rewrites emails faster. He tries to sound more current in meetings. He becomes careful about admitting confusion because confusion can look like weakness in workplaces that already reward confidence more than honesty.

European survey material points to men reporting more workplace AI use than women. That can become an advantage. It can also become a trap. If men adopt faster because they feel they must remain technically useful, the public story becomes “men are more confident with AI,” while the private reality may be closer to this: men are trying not to become disposable.

That matters.

Because when work becomes a constant audition, people do not simply innovate. They begin to manage themselves like products. They check whether their skills still sound current. They avoid looking behind. They say yes to tools they barely understand because saying no feels dangerous. They perform adaptability, even when the workplace has changed the rules without giving people time to rebuild.

Women often feel the pressure differently, and again, not as moral superiority.

Pew data has shown women are more negative than men about AI’s future impact. That is often treated as attitude. But behaviour and job structure matter more than attitude. Many administrative, clerical, coordination and communication-heavy roles are highly exposed to generative AI. These are areas where women are often strongly represented. If the first layer of AI efficiency lands on writing, scheduling, summarising, customer messages, documentation and routine office work, then women may not be “less optimistic” in some vague emotional sense. They may simply be closer to the jobs where the first cuts, redesigns and silent productivity demands show up.

A woman in an office may not say she is afraid of AI. She may just start double-checking every output. She may become the person who cleans up the AI-generated mistake. She may be told the tool will save time, while still being responsible when the tool produces nonsense. She may spend more of her day validating, correcting and smoothing the machine’s work, then be told the team has become more efficient.

That is not freedom. That is quality control with a smile.

The sharp contrast is this: men may be pushed to prove speed, while women may be more likely to carry cleanup work in AI-exposed administrative systems. One side is pressured to stay ahead. The other side is pressured to absorb the mess. Both are being told this is progress.

This is where the gender difference matters without turning it into a gender war.

The problem is not “men love AI” and “women fear AI.” The problem is that different roles create different risks. Access, job type, trust, training and power are not evenly distributed. If the conversation becomes a personality debate, we miss the real behaviour: people are adapting according to where the pressure lands.

The same thing is happening outside the workplace.

In local communities, people are no longer just reading about AI. They are reacting to the physical infrastructure behind it. A Reuters/Ipsos poll found that only about one-third of Americans support rapid AI data-centre expansion, while a majority would oppose a data centre in their own community. Many were worried about higher electricity prices. In Oregon, conflict around data-centre tax breaks has become a school-funding and local-budget issue. In Texas, reporting around water use showed weak responses from data centres to state water-use reporting requirements, leaving lawmakers without the basic visibility they need to plan.

That is not just NIMBY behaviour.

Sometimes local opposition is selfish. Sometimes it is poorly informed. But sometimes it is the only remaining brake when the official language has already decided that “development” is good and “concern” is backward.

Ordinary people notice when large companies receive tax benefits while schools, grids or public services face pressure. They notice when sustainable infrastructure arrives with unclear water data. They notice when the language of progress becomes very specific about private opportunity and very vague about public cost.

The uncomfortable truth is that AI may not destroy work in one dramatic wave. It may do something quieter and more socially corrosive.

It may narrow the first step.

Reuters reported on a Swiss jobs.ch analysis of 7.3 million job ads showing that entry-level roles were down sharply compared with the 2019–2022 average, while senior roles in AI-exposed areas grew. That does not prove a simple “AI caused everything” story. Labour markets are messy. But it points to a real danger: if AI removes routine tasks, it may also remove the training ground where people used to learn.

That is brutal for young people, career changers and people without elite networks.

“Just upskill” sounds reasonable when you already have a job, time, confidence and access. It sounds very different when you are trying to get the first chance that lets you build experience in the first place. If junior work is redesigned into senior expectations, the labour market becomes a locked door with a motivational poster on it.

This is one of the more dishonest parts of the current AI conversation. We keep telling people to adapt, but we do not always preserve the pathways that make adaptation possible.

A young worker cannot become experienced before being allowed to gain experience. A normal employee cannot become AI-ready if training is treated as personal responsibility squeezed into evenings. A local community cannot give informed consent if the important numbers are hidden in technical documents, private agreements or missing water reports.

The positive truth is also real.

AI can help. It already does help many people. Gallup has found that many employees in AI-adopting organisations report positive effects on productivity and efficiency. Some tools remove boring work. Some people with poor writing confidence can communicate better. Some small teams can move faster. Some disabled workers may gain practical support. Some data-centre operators are responding to pressure with better cooling systems, including designs that reduce or avoid water use for normal cooling.

That matters, because a serious critique should not pretend the benefits are fake.

The point is not to stop useful tools. The point is to stop pretending that usefulness cancels out the bill.

If AI gives a worker two hours back, good. If AI simply raises the expected output and calls the exhaustion efficiency, then the human did not gain freedom. The system gained capacity. If a data centre brings local jobs, fine. If it also receives major tax benefits, consumes scarce grid capacity and leaves residents unclear about water and energy impacts, then people have a right to ask whether the deal is actually public progress or private extraction with better branding.

This is where the harmful narrative hides.

It does not always sound cruel. Often it sounds caring, modern and responsible. “We need to prepare people for the future.” “We need to stay competitive.” “We need to embrace innovation.” “We need to be flexible.” Those sentences can be true. They become harmful when they erase the practical question underneath: who is being asked to carry the transition, and what support do they actually get?

Adaptation is not neutral when the cost is not shared.

A company can say AI is a tool, not a replacement, while quietly expecting fewer people to do more. A politician can say data centres are part of national competitiveness while local residents worry about power bills. A manager can say “experiment with AI” while never defining data rules, accountability, training time or what happens when the tool is wrong. A school system can tell young people to prepare for the future while the first rung of the job ladder disappears.

The contrast matters because it shows the gap between public language and private behaviour.

Publicly, everyone is pro-future. Privately, people are saving receipts. They are learning tools in silence. They are avoiding looking replaceable. They are checking job ads and noticing “entry-level” roles asking for senior skills. They are asking whether a new data centre will raise bills or reduce local control. They are using AI to save time, then spending the saved time keeping up with the next demand.

That is the part RD should care about.

Not panic. Not hype. The lived cost.

So what does everyday self-defence look like here?

Not guru nonsense. Not “master AI in 30 days.” Not “change your mindset and win the future.” A tired person does not need another performance project. They need a clearer way to see where the work has been moved.

At work, when a new AI tool is introduced, ask for the boring things: What is it for? Who is responsible when it is wrong? What data can be entered? Is training time actually work time? What will be measured before and after? Which tasks should not be automated? Those questions are not resistance. They are adult boundaries.

In your own work life, do not confuse shame with strategy. If you feel behind, name the actual condition: a new access barrier has appeared. Then choose one small practical move. Learn one tool that touches your real work. Save examples of what you can do. Keep evidence of your output. Do not try to become a full-time AI influencer just to remain employable.

In your local community, when a data-centre project is sold as green growth, ask for specifics before slogans: electricity demand, water use, tax agreements, local jobs, noise, backup power, grid impact, deadlines and who is accountable if promises fail. A project that is truly good for the public should be able to survive public questions.

In private life, use AI without worshipping it. Let it help with drafts, structure, translation, planning or admin if it saves you real energy. But keep your judgment. Do not let convenience train you into passivity. A tool that helps you think is useful. A tool that slowly replaces your responsibility is not.

The direction is simple: more clarity, less hidden work. More documentation, less trust theatre. More training inside paid time, less private panic. More local voice before infrastructure is locked in. More dignity in the meeting between human beings and systems.

AI can be part of the future. But ordinary people should not be forced to become the shock absorber for every side effect while being told to smile and adapt.

The question to carry with you is: who has moved the work, risk or bill onto me here — and where is the real support?

Sources and how they affect everyday life

International Energy Agency — data-centre electricity demand and AI energy pressure
Link: https://www.iea.org/reports/key-questions-on-energy-and-ai/executive-summary
Summary: The IEA projects major growth in electricity demand from data centres, with AI-focused infrastructure driving much of the increase.
Everyday impact: This turns AI from a software issue into a household and grid issue. More demand can affect energy planning, local infrastructure and eventually prices.

Energinet — temporary pause on new grid connections in Denmark
Link: https://en.energinet.dk/about-our-news/news/2026/temporary-pause/
Summary: Energinet announced a temporary pause while handling a very large queue of new electricity consumption projects.
Everyday impact: Shows that the Danish grid is not an infinite background system. Ordinary people may feel this through delays, prioritisation conflicts and public debate over who gets capacity first.

Reuters — EU plans energy-efficiency standards for data centres
Link: https://www.reuters.com/business/energy/eu-plans-energy-standards-data-centres-amid-concerns-over-soaring-power-use-2026-06-03/
Summary: EU officials are looking at minimum standards because data-centre energy use is becoming too large to ignore.
Everyday impact: Confirms that data centres are moving from tech policy into energy policy. Citizens should expect more political fights over electricity, standards and accountability.

Reuters — US energy regulator pushes grid overhaul for data-centre power rules
Link: https://www.reuters.com/business/energy/top-us-energy-regulator-pushes-grids-overhaul-data-center-power-rules-2026-06-18/
Summary: US regulators are asking how large new electricity users, including data centres, can connect without weakening reliability or shifting costs onto other customers.
Everyday impact: This is the household-bill question. If rules are weak, normal customers can end up paying for infrastructure built for massive private demand.

Reuters/Ipsos — Americans wary of AI data-centre expansion
Link: https://www.reuters.com/world/us/americans-wary-ai-driven-data-center-boom-reutersipsos-poll-shows-2026-06-11/
Summary: The poll found strong public concern over local data-centre expansion, especially around electricity prices and local siting.
Everyday impact: Shows that local opposition is not just abstract climate concern. People are reacting to concrete fears: bills, land, noise, water and loss of local control.

Axios Portland — Oregon data-centre backlash and tax breaks
Link: https://www.axios.com/local/portland/2026/06/24/data-center-backlash-ai-expansion-oregon
Summary: Covers conflict around data centres, tax benefits and local public finance in Hillsboro, Oregon.
Everyday impact: Makes the class issue visible. When large companies get favourable deals, residents may ask what happens to schools, services and local budgets.

Houston Chronicle — Texas data centres and water-use reporting
Link: https://www.houstonchronicle.com/politics/texas/article/data-center-water-usage-reports-required-texas-22303230.php
Summary: Reporting showed weak response rates to water-use reporting requirements, leaving lawmakers with limited visibility.
Everyday impact: If communities cannot see water use, they cannot judge risk. Lack of transparency turns “trust us” into a public burden.

Reuters — Swiss study on fewer junior roles in AI-exposed jobs
Link: https://www.reuters.com/business/fewer-job-offers-junior-roles-due-ai-swiss-study-shows-2026-06-24/
Summary: Reports on a jobs.ch analysis showing lower entry-level postings compared with earlier years, while senior roles in AI-exposed fields increased.
Everyday impact: This affects young workers and career changers directly. If first-step jobs shrink, “just gain experience” becomes a closed loop.

Gallup — AI adoption and workforce changes
Link: https://www.gallup.com/workplace/704225/rising-adoption-spurs-workforce-changes.aspx
Summary: Gallup finds many employees in AI-adopting organisations report positive productivity and efficiency effects, while also showing that AI is changing workforce expectations.
Everyday impact: AI is not fake hype. It can help. But if productivity gains become higher output expectations without recovery or training, workers carry the cost.

Pew Research Center — gender differences in views of AI
Link: https://www.pewresearch.org/chart/women-are-more-negative-about-the-future-of-ai-than-men/
Summary: Pew shows women are more negative than men about AI’s future effects.
Everyday impact: This should not be reduced to attitude or confidence. It may reflect different job exposure, different trust levels and different experiences with technology-driven systems.

European Commission — AI adoption divide among workers
Link: https://economy-finance.ec.europa.eu/economic-forecast-and-surveys/economic-forecasts/spring-2026-economic-forecast-slowdown-growth-energy-shock-drives-inflation/ai-adoption-divide-who-benefits-who-doesnt-and-what-it-means-workers_en
Summary: EU material points to differences in who adopts and benefits from AI, including by gender, education and type of work.
Everyday impact: Shows that AI opportunity is not evenly distributed. People with more access, confidence and training can move ahead faster, while others are told to adapt from behind.

Axios — Microsoft and lower-water-use AI data centres
Link: https://www.axios.com/2026/06/24/microsoft-lower-water-use-ai
Summary: Covers Microsoft’s move toward data-centre cooling designs intended to reduce or avoid water use during normal operation.
Everyday impact: This is the useful positive signal. Public pressure can force technical improvement. But water-saving claims still need to be read alongside electricity demand and local grid impact.