You're pledging to donate if the project hits its minimum goal and gets approved. If not, your funds will be returned.
I produce accessible short-form videos about AI risks for my German-speaking audiences on Instagram, YouTube and TikTok. Within 14 months, my channels have grown to over 90,000 followers. In September 2026 alone, my content reached about 2.5 million views as well as 200,000 interactions on Instagram. Among my viewers are numerous members of the German parliament and other influential people.
So far, I do all of this work by myself on the side. I work on this during the evenings and on weekends. I have a full-time job as a "program manager for democracy promotion" in public administration for the city of wolfsburg.
The biggest strength of this project is: German-language content on AI safety is scarce. There is no other large German-language channel that focuses on catastrophic AI risks to my knowlegde. Most accessible material on loss of control, misaligned systems and AI governance exists only in English. Some channels cover these topics on the side, but not regularly. The German-speaking audience is therefore severely underserved (even though the German public is generally very politically engaged and Germany plays a central role in European policymaking).
I am asking for 12 months of funding to work full-time on my existing channels. This will allow me to publish more often, produce long-form videos, and most importantly: scale the channel.
It will also give me the time to build two sources of long-term income to make sure, my work can sustain after the duration of this grant. I want to establish myself as a keynote speaker and find sponsorships that actually fit my channel.
My overarching goal is to make sure, that the broad german public is aware of the AI safety discourse. Most of Germans have no accessible source for AI safety in their own language. Especially on the social media feeds where (especially younger people) get their information. I want to close that gap.
Over the 12 months, I plan to:
Produce at least three high-production short-form videos per week. On top of that, I want to create at least two longer explainers on YouTube a month as well.
Cover risks from advanced AI (loss of control, misaligned systems, concentration of power, and rogue AI agents from real world accidents), AI governance and regulation, as well as everyday risks like sycophancy and hallucinations in chatbots. This also includes publishing the accompanying videos with acces to all sources on my website (https://niklastitgemeyer.de/quellen.html).
Release a storytelling podcast series on whether advanced AI can be controlled. In this podcast series I want to dive deep into questions like: What are the philosophical arguments behind the doomsday scenario? What is misalignment? What problems does an "off-switch" have? I have applied separately to the Future of Life Institute for funding for this podcast. The decision is currently pending. If that application succeeds, the podcast will be covered by FLI, and the funding raised here will go entirely towards the other activities listed on this page.
Establish myself as a keynote speaker on AI risks in germany. This is a very important subgoal, so I can sustain my work after the funding period ends. My goal is to secure at least 20 keynotes by the end of 2027.
I already do much of this every single week. The funding, however, would allow me to do it at a much higher frequency.
I will track reach, follower growth and engagement across all platforms and post regular progress updates on this page. My target for the end of the grant is 200,000 followers on Instagram. My Instagram account has already grown to 66,000 today, and the growth is steady. That's why I am confident, that this will work.
Funding goal: $100,000 for 12 months Minimum funding: $70,000
$82,000: salary for full-time work, including income tax, mandatory health and long-term care insurance, and pension contributions as a self-employed person in Germany.
$3,000: video production equipment, software and conference attendance.
$15,000: freelance video editors, better equipment, outreach to new audiences and keynote clients.
$70,000 is the lowest amount at which I could leave my current job and still cover my living costs, including taxes and social insurance. At this level, I would cover only a reduced salary and no equipment costs. $85,000 covers full-time work and basic production costs. Anything above that would significantly increase the quality and reach of the project.
I work on this alone. I research, script, film, edit and publish all content myself. My experience covers several areas:
Science communication: I have built an audience of over 90,000 followers across Instagram, YouTube and TikTok with content on AI risks. My videos reach millions of people every month. Since July 2026, I have also been working as a keynote speaker on AI risks and have been booked for more than five keynotes so far.
Academic background: I am a sociologist (M.A.) specialised in the sociology of technology.
Journalism: I have worked as a freelance reporter for the regional newspaper Neue Westfälische for four years and at the Bielefeld studio of WDR (one of Germany's largest public broadcasters) for one year.
Content marketing: I spent more than a year at a content marketing agency, where I have gained practical experience in creating content for digital platforms and reaching specific audiences.
Democracy work: Since 2024, I have been working full-time as a programme manager for democracy promotion in public administration, where I manage the local Partnership for Democracy. In this role, I have supported more than 100 projects, which together have reached over 100,000 people, and I have given more than 30 workshops and talks. My earlier work in civic education includes teaching democracy and media literacy in prisons.
AI training: I have completed Harvard's "The Science and Implications of Generative AI", Anthropic's "AI Fluency" course and several further courses on LLMs, responsible AI and AI ethics.
This combination of science communication, journalism, content marketing and democracy work allows me to explain complex AI risks in a way that reaches people who would otherwise never engage with the topic.
Examples of my work:
Instagram: https://www.instagram.com/niklastitgemeyer/
Website: https://niklastitgemeyer.de/en/
Viral video (2 million views) explaining the Hugging Face incident with a hand-drawn diagram: https://www.instagram.com/niklastitgemeyer/reel/Dc3WEVLOt2Z/
Video explaining the core arguments for existential risk from AI: https://www.instagram.com/niklastitgemeyer/reel/DdOmozhOKlc/
Viral carousel on studies about hallucinations and sycophancy in chatbots: https://www.instagram.com/niklastitgemeyer/p/Dc26KOaDllI/
Viral carousel on the risks of developing AGI for jobs and society: https://www.instagram.com/niklastitgemeyer/p/DY1Omvyjr6K/
Most likely causes:
The keynote business grows too slowly. If I cannot build enough keynote income (or alternative income streams such as aligned sponsorships) during the 12 months, I will not be able to continue full-time after the grant. This is why building it will be my second priority, right after creating engaging content.
Growth falls short of my target. I may not fully reach 200,000 followers on Instagram by the end of the grant. Even then, this would not mean the project has failed. My channel is already growing steadily with the approach I use today. Based on my growth so far, I expect this growth to continue and every additional video still reaches tens to hundreds of thousands of German speakers who would otherwise hear little about AI safety.
Most likely outcome if it fails: I return to employment and continue producing AI safety content in my spare time, as I do now. The work would go on, but at a lower frequency and with less depth. The content already published stays online, and the audience, skills and network I build during the grant remain.
$0. All of my AI safety content so far has been unpaid. I have not received any grants, donations or sponsorships for it. Pending applications related to this project:
Coefficient Giving (September 2026): funding for this same project. Decision pending. I will not accept double funding for the same work. If this project reaches its minimum funding here first, I will withdraw my application at Coefficient Giving. If Coefficient Giving commits first, I will close this project and return all funds raised here. If they fund only part of it, I will lower my goal here by the amount they provide.
Future of Life Institute (submitted on 30 September 2026): $12,000 for the podcast series described above. Decision pending.