You can try Medienkundig before creating an account. The current introduction offers three short exercises, then lets you decide whether to save your progress. That order responds to a question we encountered during the project: how can someone judge a learning product before trusting it with an account?
Jacqueline Lehmann and I developed Medienkundig through our media-design studies. It explores how a learning application can support older adults in dealing with digital deception. Open the app to try the German-language exercises.
Start by taking a web address apart
In the entry flow checked for this article, the first task presents a web address and asks the learner to place its components in the correct gaps. You select a term, choose its position and submit the answer. The app responds, then explains the domain, the encrypted connection and the path to a page.
Choosing the wrong component gives the learner something specific to revisit. The progress indicator moves through the three introductory tasks.
The accompanying website describes further work with topics, repetition and a reference dictionary. Its methodology page explains the training approach.
Put the latest crime figures in the right year
Germany's Federal Criminal Police Office published its 2025 cybercrime findings in May 2026. The release reports approximately 335,000 recorded cases of cybercrime in the narrower statistical category for 2025, including 207,888 attributed to locations abroad or unknown locations. It also warns of substantial unrecorded crime. This category does not cover every kind of fraud committed using the internet. BKA release, 12 May 2026.
The reporting year is 2025. A complete 2026 total is not yet available.
The 2026 Cybersecurity Monitor asks a different question. In its January survey, 32 percent said they had never taken concrete steps to check whether an image or video was truthful. Survey responses describe reported behaviour; they aren't police case counts. BSI and ProPK: CyMon 2026.
Forty-three tried it. One registered.
The bachelor's thesis version dated 14 August 2026 documents 32 usable responses to an open survey of fraud experiences, conducted in July, and three written expert interviews. Those accounts informed the scenarios. The sample was not designed to estimate how common fraud is across Germany.
The thesis also records an early, unmoderated field test from 2 to 5 August. The application reached 50 people through personal recommendations. Forty-three completed the introductory questions; one registered afterwards. The accompanying feedback described uncertainty around account creation and the flow of the product.
The drop at registration was hard to overlook. The short test tells us about that entry flow; it cannot establish lasting learning or fewer fraud incidents. Today's introduction explicitly says that an account is unnecessary at first and that saving progress is a later choice. A further test would be needed to assess whether this version resolves the earlier difficulty.
Why use exercises at all?
The design draws on short learning steps and repeated recall. A systematic review by Agarwal, Nunes and Blunt found benefits from retrieval practice across school and classroom research. Those findings support investigating the method; they do not establish an effect for Medienkundig or for our target population. Educational Psychology Review, 2021.
The OECD's 2025 brief on digital skills for older adults also argues for training that reflects learners' circumstances. Flexible pacing and support deserve attention when designing an entry experience. OECD: Digital skills for seniors.
In a 2025 study based on interviews with 34 digital educators in Ireland, participants described the importance of individual guidance and hands-on learning, alongside barriers such as outdated devices and limited support. I would include those conditions in any further evaluation of our app.
Germany’s DigitalPakt Alter is working with ten pilot municipalities in 2026 to develop structures for older adults’ digital participation. A supported session at a community centre could offer a setting in which to test Medienkundig alongside someone available to help.
For the first group session, ask a facilitator to observe one learner's route through an exercise: opening it, making a choice, reading the explanation and describing what they would check next time. Keep two separate notes: help with operating the device, and help with understanding the task. They lead to different changes in the next session.
Allow people to use a familiar device where possible. If opening the app takes the available time, that is useful information about the entry experience. A later meeting can revisit the explanation with a different example. This is a suggested facilitation approach, not a tested programme duration.
Speaking about new AI learning venues on 25 August 2026, BAGSO chair Dr Regina Görner connects digital participation with social justice and calls for places where older people can try technology and ask questions. That access question is relevant before a facilitator starts assessing answers.
You can try Medienkundig here. For a supported group trial, visit Medienkundig for education providers and municipalities. Foundations can also discuss a partnership or funded pilot. These German-language pages explain the current offer; mention the learners, setting and available devices when enquiring.
Research and app entry checked on 10 September 2026. Project source: bachelor's thesis, 14 August 2026 version, sections 4.3, 5.3 and 7.2.5, especially pp. 56–57 and 126–127. Cover: actual Medienkundig app screen showing the web-address exercise.
