How our phones choose the political class instead of us

Humanity tends to tackle a risk after the risk already affected people

One could argue that people are either not doing very well in proactively mitigating risks, or they overexaggerate, and sometimes very late. For example, television became popularly adopted in the US during the early 1950s. In European countries, adoption lagged slightly, gaining mass popularity in the late 1950s to early 1960s.

In 1974, despite unclear evidence, the US had its first anti-subliminal TV messages law set. It took the EU 33 more years to follow the lead.

We can see this pattern replicating in many areas, but as we transitioned from TV to the internet as our source of getting informed, internet regulations lagged behind quite a bit.

Pornography became available to the wider public in 1994–1995 with the launch of the Mosaic and Netscape browsers, making explicit content instantly viewable to the public, leading to the first commercial adult empires. Only in 1996–1998, in response to the US Communications Decency Act (1996), sites implemented the now-classic “I am 18” button. This was a self-attestation “legal shield” with no actual proof required. Only in Jan 2023 Louisiana became the first Western jurisdiction to mandate government-ID checks for access. This triggered a rapid Western expansion, with the UK (2025), France (2024), and Australia (2026) implementing “hard” verification via digital IDs and facial age estimation.

Today, nearly everyone has access to the internet, including young children. However, unless there is additional software used, they still have access by default in the majority of the world to parts of the internet that are now proved to affect their brains and how they are growing (Sources: Digital Media Use and Child Health and Development: A Systematic Review and Meta-Analysis, published in JAMA Pediatrics, March 2026; The Effects of Digital Addiction on Brain Function and Structure of Children and Adolescents, published in MDPI Health, Updated 2024/2025)

UX used as a weapon

The Online manipulation is rarely about dialogue; it is about design. “Dark Patterns” are standardized UI/UX tactics specifically engineered to subvert user intent.

Going forward, you will have the chance to review a few ways internet users are manipulated. Some of them will feel familiar. Fortunately, I would argue, the EU played the unpopular card and already improved the situation of its citizens asking to have some of these behaviors regulated and to avoid unethical or harmful web platforms’ behaviors.

Becoming aware of these methods is the first step towards building resilience against them.

The “Roach Motel” & Subscription Inertia: This design makes entry into a commitment effortless while making leaving a platform nearly impossible. In late 2025, a major European streaming service was fined under the Digital Services Act (DSA) because while account creation took 12 seconds, cancellation required a 15-minute “retention chat” and navigating five hidden sub-menus.
Confirmshaming: This is the practice of phrasing an “opt-out” choice to trigger a shame response. If a user tries to decline data tracking in 2026, they are often met with a button reading: “No, I prefer to remain uninformed and pay full price.” This triggers a micro-social rejection response, nudging the user toward a “Yes” they never intended.
Online retailers successfully manipulate the human perception of value through Anchoring. The human brain is actually pretty bad at knowing what something is worth. It needs a “price anchor” to compare and can only judge relative value. Once a product or service is shown at a specific price, “special limited offers” for a smaller price create a sense of urgency for an immediate purchase.
The Decoy Effect is another way of anchoring.

The Setup: Imagine you want file storage in the cloud and you get the following offers from a cloud provider:

Option A: 50GB for $2
Option B: 2TB for $10

Most people would take the $2 deal. So, the company adds:

Option C (The Decoy): 1TB for $9.

The Result: Suddenly, the $10 deal looks like a steal! You end up spending $10 instead of $2 because the “Decoy” made you feel like you were “winning.”

Tier C exists solely to make Tier B look like an exceptional “deal.”

While commercial practices are easiest to spot due to their history in physical commerce, other online manipulation techniques have changed the world as we know it. Look no further than Brexit.

The most flagrant example of manipulation is the evolution of the OCEAN personality model (Openness, Conscientiousness, Extraversion, Agreeableness, and Neuroticism). This is a big one. Apps use your data to figure out your personality type using the OCEAN model.

In the 2025/2026 European election cycles, campaigns utilized AI-driven micro-targeting to identify “High Neuroticism” voters.

The Tactic: Instead of policy debates, these individuals were served hyper-targeted “fear-porn”—images of civil unrest or economic collapse—to bypass logic and trigger a primal security response.
The Result: A landmark study in Science (2025) confirmed that reranking a social media feed to prioritize maliciously critical content can change one’s political opinions as much as three years of real-life experience. People are seeking information and entertainment online, but they get re-programmed instead.

The €120 million fine issued to X (formerly Twitter) in December 2025 proved once again that deceptive designs (such as the “Blue Checkmark” system) facilitated massive impersonation scams and political deepfakes. This has created a “Trust Deficit” as the Flash Eurobarometer 541: Democracy, published by the European Commission (2024) shows that 78% of EU citizens are concerned about disinformation and “cyber-manipulation” influencing people’s voting decisions.

However, despite having these and many other practices documented or studied and despite even having elections canceled due to online practices (i.e. the Romanian presidential elections), we still do not have our citizens properly protected against malicious online practices.

What does the future reserve?

We like to think we’re in the driver’s seat. We choose what to buy, who to follow, and what to believe, right?

Well, in 2026, the “driver’s seat” is actually a high-tech simulator. We may still believe that we are turning the wheel, but the road is being built in real-time by an algorithm that knows our weaknesses better than our best friends doe. This is called Behavioral Engineering.

It’s all depending on us.

According to The EEAS Report on Foreign Information Manipulation (FIMI), there is a 65% increase in the use of AI tools to produce manipulative content compared to the previous cycle. The 2026 report documents identify the “Information Space” as a new frontline in national defense, noting that manipulation is no longer just “fake news” but a coordinated attempt to undermine Cognitive Sovereignty, with applications across the society. This can, of course, evolve into worse and worse status quos.

However, let’s bring a bit of optimism to the table. How could we use today’s technologies to make the world better?

While humanity is historically bad at tackling risks in real-time and in the western world we tend to be rather profit-driven, humanity also showed throughout time that it can ingeniously make use of the inventions available in the market to make life and society better.

Imagine a school where a goldfish, a monkey, and an elephant are all given the exact same final exam: “Climb that tree.”

For decades, that’s been the European education system. It’s standardized, rigid, and frankly a bit soul-crushing for the “fish” and “elephants” of the world. But in 2026, a new player has entered the game: The Algorithm. The problem? Right now, the algorithm seems to be used solely to increase margins and sell things. But it’s actually the same technology that could finally enable the fish to swim qnd the elephants to happily roam free.

Apps already use your data to build a Psychographic Profile (the “OCEAN” model). They know if you’re anxious, adventurous, or easily distracted.

The Manipulation: If you’re a “fish” who struggles with focus, the algorithm doesn’t help you swim; it builds a digital “bubble” of 15-second cat videos to keep you stuck.
The Result: We see Confirmshaming and Decoy Effects—tactics that use your specific personality to nudge you into staying online longer than you planned.

Here’s a plot twist: That same “creepy” data that knows how to distract you also knows exactly how you learn. That data can feed an AI tutor.

Personalized Learning: Instead of the “Climb the Tree” exam, AI can now act as a 1 on 1 tutor for every child in Europe. If a child is a visual learner who loves space, the AI doesn’t teach math with boring black-and-white worksheets but it teaches calculus through rocket trajectory simulations. Of course, the AI doesn’t have to be the actual teacher, but a personalized educational program could be handed over to teachers to ensure we focus on and leverage the unique qualities of each pupil.
Ending the “Fish”/industrial era education system design: For the first time, we can stop evaluating every kid by the same narrow metric. The AI adapts to the child’s pace, mood, and interests.

Summary

Digital Awareness in 2026 is about realizing that data is power. As consumers, we can demonize corporations for owning that data and get manipulated, or we can leverage their findings and the wonders they have produced to protect citizens and even drastically improve their lives. If we use that data for personalized education, we finally get a system where the fish is allowed to be the best swimmer in the world, rather than a “failure” at climbing trees.

What other ways of using AI for the better do you have in mind?