Autonomous Intelligence Forum #2
Wednesday, 11 November 2026. Doors 14:00, programme 14:30 to 18:30, then buffet and conversation. Registration via Luma, 70 seats.
Signing authority goes to people you know and trust. AI agents are being handed it in passing. They write quotes, answer customers and book invoices. What goes on inside them, nobody sees.
This afternoon walks the whole stretch: from what these systems can really do today to who answers for it in the end. Knowing them is what tells you which project will hold up in your own house and which one fails the first audit.
Made for decision makers who are not in the technology themselves: from the Mittelstand, from corporates, law firms, medical practices and public administration. Short talks, questions from the floor, conversation at eye level. No endless panels. At the end the question: may the agent sign?
What works and what applies
Christoph Mütze, tool developer and designer from Hamburg, stands at the forward end. He works where generative AI meets real-time graphics. He tries new tools before others have heard of them, and where they fall short he builds his own. From him comes the answer to what actually works right now and what merely looks good.
Dr. Pascal Wichmann, Head of AI at Caps & Collars, takes the question that follows immediately: how can you tell at all that a machine wrote, drew or spoke something? And how reliable is that detection when it matters? The question has a deadline. Since August the AI Act has required providers of generative AI to mark their output as machine generated in a machine-readable way. For systems that were already on the market, the transition period ends on 2 December.
Alois Krtil, founder and managing director of ARIC, places where we stand on agentic systems, generative AI and automation. Alongside that, what is taking shape in Europe: regulation, new frameworks, sandboxing, security. What already applies, what is still a draft and what to prepare for.
What holds up in your own house
Nikolai Bockholt stands where an experiment turns into an operation. Ten years at Google, plus AI in regulated healthcare, with data that falls under Article 9 GDPR. Today Head of AI at Carlsquare, an international M&A advisory. He builds himself and recommends nothing he has not seen in production.
Oke Göttlich, president of FC St. Pauli and shareholder in KI:KUBE, takes the question to where the machine physically stands. KI:KUBE builds AI infrastructure for your own building instead of someone else's cloud: a device in the server rack, on an ordinary socket. The reason is the same one this afternoon is about: confidentiality, evidence, and the question of who can switch off your access. Since June that is no longer theory.
Who answers for it
Prof. Dr. Frank Passing stands at the other end, in regulated operations. His questions: who checks, who is liable and what has to be evidenced once autonomous systems arrive there. He teaches digital transformation at IU and builds agents with Fairfield & Archer that screen regulatory changes and route them to the right place in the company.
Closing round: may the agent sign? Nikolai Bockholt for operations, Oke Göttlich for infrastructure and Prof. Dr. Frank Passing for liability each answer in one minute. Oliver Rößling moderates.
DEEPTECH Campus Circle Hamburg
Become a member to attend future exclusive events.