Finnish Center
for Safe AI Tutke
Tutke promotes safe artificial intelligence
Tutke is an interdisciplinary non-profit organisation. We promote scientific understanding of the safe development of artificial intelligence and support research and regulation leading to it.
Artificial intelligence has progressed enormously. Today AI can converse fluently, solve mathematical problems, write code, and create art and realistic video. All this has become possible in the span of a couple years.
Leading artificial intelligence research centers and companies are racing to develop general artificial intelligence that surpasses human abilities in all tasks. Predicting when and if this will be achieved is difficult, but experts believe it is likely to happen within this century. When machines begin to surpass humans, such development may accelerate even more.
The impact of such general artificial intelligence would be societally momentous, similar to or greater than the industrial revolution in the 19th century. If we navigate this change well, it could bring unprecedented abundance and solutions to humanity's largest problems. If we fail, it could lead to issues like extreme concentration of power. In the worst case, we create inhuman beings that surpass our capacity to control them, and lose the ability to decide our own future.
It is important that instead of only charging towards these volatile consequences, we act early to understand and wisely manage the societal risks of such transformative technology. This requires wide-ranging cooperation between researchers, technology companies, and decision-makers.
Network members
Kaj Sotala
Research
Kaj Sotala started work on AI safety in 2012 and has since authored several papers on the subject. He has been a Research Fellow at the Machine Intelligence Research Institute and a Researcher at the Foundational Research Institute. Kaj has degrees in Cognitive Sciences and Computer Science from University of Helsinki.
Onni Aarne
Research
Onni Aarne researches compute governance and has done work for the Institute for AI Policy and Strategy and Rethink Priorities, where he worked in the AI Governance and Strategy team. He has a degree in Computer Science and Data Science from University of Helsinki.
Rudolf Laine
Research
Rudolf was the technical team manager on the Situational Awareness Dataset benchmark, published in NeurIPS and used in OpenAI's pre-deployment testing. He has completed a Master's degree in Computer Science at University of Cambridge, and has studied in a machine learning PhD program at University of Oxford.
Olli Järviniemi
Research
Olli Järviniemi is researching AI safety with UK AI Security Institute, and has previously completed a PhD in Mathematics at University of Turku. He specializes in AI control and deceptive alignment.
Anna Katariina Wisakanto
Research
Anna Katariina Wisakanto is a risks from global systems and AI safety researcher. She currently works for Center for AI Risk Management & Alignment. Previously, she worked as a visiting scholar at the Leverhulme Centre for the Future of Intelligence, University of Cambridge.
Patrik Gayer
Policy & Communications
Patrik Gayer is the Director of Government Relations at AMD (formerly Silo AI), and has previously worked as a AI & Data Policy Lead at Tesla, and as a Director at Tekir. Patrik has a degree in Political Science from University of Helsinki and in Public Administration from Harvard University.
Joonas Ollila
Strategy
Joonas Ollila is a technologist leader who has been the CEO of optimization firm Weoptit and the CEO of Vuorox, amongst other roles. He has a degree in Systems Analysis and Operations Research from Aalto University.
Joel Pyykkö
Research
Joel Pyykkö is an AI safety researcher with a Doctoral degree in Computer Science from University of Helsinki.
Suvi Auvinen
Strategy & communications
Suvi Auvinen is a journalist, activist, and writer. She leads responsible AI strategy at Ellun Kanat, and has a Bachelor of Arts from Turku University of Applied Sciences and a Master of Arts from University of the Arts Helsinki.
Santeri Koivula
Policy & governance
Santeri Koivula is a Master's student in Science, Technology and Policy at ETH Zürich. He has researched AI risks and EU AI policy at Future of Life Institute, and interned at Demos Helsinki and CERN. Santeri was a Pro Forecaster for INFER Public (RAND Forecasting Initiative), and is part of an AI expert panel at Forecasting Research Institute.
Dmitrii Gusev
University Coordinator
Dmitrii Gusev studies Computer Science at Aalto University. He is the lead organizer of the technical reading group "LLMs and Alignment", and he co-organized the 2025 Finnish Alignment Engineering Bootcamp.
Operations
Mikko Hemmi
Director
Mikko Hemmi has a Master's degree in Computer Science from Aalto University. He has previously worked in investment banking and software, and has advised many leading Nordic enterprises, startups, and public sector organizations as the CEO of data & AI consultancy Oikoa.
Santeri Tani
Operative Director
Santeri Tani has studied Computer Science at University of Helsinki. Previously, he has worked in robotics, machine vision, and health tech. He is also advancing forecasting and preparedness capabilities through the RAND Forecasting Initiative.