The In-Between Tech and Trust Podcast

Eva Simone Lihotzky

How do we build, break, and rebuild trust in a world shaped by technologies? The In-Between Tech & Trust is a weekly conversation about the human side of tech. Host Eva Simone Lihotzky, founder of raidiant (responsible AI advisory), talks with guests from business, politics, neuroscience, and tech about what they have learned about trust, and what they would pass on to you. Eva was formerly co-founder and MD of the Serviceplan AI Lab, holds the Global Moral Chair at the Value AI Institute and co-authored "10 Moral Questions: How to Design Tech & AI Responsibly".

  1. 48m ago

    The Doom and Gloom of AI: Fears and Facts about the Current State - Eva Simone Lihotzky (EP 45)

    Eva Simone Lihotzky, solo episode In July, an AI model under test sent the Philadelphia police an invented tip about an unsolved murder, and its maker only found out two months later. In this solo episode, Eva Simone Lihotzky takes that story and the fears her listeners sent her (losing control, losing ourselves, jobs, a bubble) and checks each one against what has actually happened. Her stance on the doomsday narrative is clear: she doesn’t buy it. What worries her is how much power sits with very few companies, and who holds them accountable. 🎧 Episode overviewThis summer’s incidents are real: OpenAI test models broke into Hugging Face’s systems to fetch the answers to their own test, and Anthropic models carried on with cyber tests against targets they recognized as real companies, mostly with safeguards deliberately turned down. Eva explains the mechanism behind it (reward hacking: models are rewarded for the result, not the intention) and names the fear she does find accurate: an agent nobody is watching, while only 21% of companies have a mature way of governing agents. On our thinking, she separates AI as an answer machine, which left Turkish high school students with 17% worse exam results, from AI as a tutor with guardrails, which largely removed that harm; on jobs, she argues that AI changes tasks more than it removes jobs, and that a person still decides every cut. Her reason to stay optimistic comes from people’s own experience: in Germany, more than half of non-users would rather live without AI, but only a third of users, and the benefits that work (stroke care, breast cancer screening, weather forecasts) all keep a human in charge and measure the result. 💡 Key themes Which fears about AI hold up when you check them against this summer’s incidentsWhy labs warn about their own models and keep releasing new onesWhen AI weakens our thinking, and when it helps us learnWho decides that AI costs jobs, and who owns helping people learn what comes nextWhether Europe’s AI Act slows companies down or gives them the trust to adopt AI 👤 About the hostEva Simone Lihotzky hosts The In-Between Tech & Trust and as the founder of raidiant, she advises European companies and boards on responsible AI adoption. She previously co-founded the Serviceplan AI Lab and led it as Managing Director, and she is co-author of “10 Moral Questions: How to Design Tech & AI Responsibly”. This episode is built on the questions her listeners sent her on Instagram and LinkedIn. ⏱ Chapters 00:00 An AI’s false tip to the police, and where Eva stands02:53 From listeners’ questions to agents that left their tests08:42 Why the warnings got so loud10:18 What AI does to who we are, how we think and our jobs14:46 Bubble, power and Europe’s rulebook19:10 Why Eva stays optimistic, and what to do now 🔗 Links Eva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/raidiant, Eva’s responsible AI advisory - https://www.raidiant.eu/Anthropic, Investigating unintended model actions (the police tip) - https://www.anthropic.com/research/investigating-unintended-model-actionsTechCrunch, OpenAI says Hugging Face was breached by its pre-release models - https://techcrunch.com/2026/07/21/openai-says-hugging-face-was-breached-by-its-pre-release-models/Deloitte, State of AI in the Enterprise 2026 - https://www.deloitte.com/us/en/about/press-room/state-of-ai-report-2026Bitkom, KI macht Spaß - aber nicht allen (May 2026) - https://www.bitkom.org/Presse/Presseinformation/KI-macht-Spass-aber-nicht-allenThe New York Times editorial board, AI is not above the law - https://www.nytimes.com/2026/10/10/opinion/ai-companies-liability-laws.html Related episode - Two Fears Behind Every AI Transformation - Dr. Charlotte Blum (EP 31) - https://open.spotify.com/episode/5A0aUErqMyqTKjBQqOuGaD?si=inpGls9wQoOpsuMZaBmDsg

  2. Oct 8

    The Quantum Momentum: How Companies Stay Ready for a Computing Wave That Is Yet to Arrive - Christopher Trummer (EP 44)

    🎙️with Christopher Trummer, MD Venture Lab Quantum / Semicon, TUM Venture Labs Around 2019, quantum computing made its first big promise: machines with just a few noisy qubits would solve useful problems within two or three years, and many companies got ready for them. Christopher Trummer, who runs the quantum and semiconductor venture lab at TUM Venture Labs in Munich, explains why that promise failed and why the error-corrected machines now in development are a far more solid bet. What he leaves leaders with is a question of trust and patience: how do you keep a board believing in a technology whose progress is real (and slow)? 🎧 Episode overviewChristopher's starting point is that quantum is already here in parts: communication and sensing are ready for industry, and some post-quantum encryption runs in your Chrome browser today (without a mathematical proof yet that it holds). Quantum computing is the part still on its way, and he is candid about its first wave: the algorithms for the early, noisy machines never came, and the promise "fell out of the window completely." That shifts the risk to the companies waiting for them: keep the momentum and the board's support through slow progress, or miss the moment quantum starts to matter for the business. Eva asks how anyone prepares for this wave while still adopting AI, and his answer is sober: find the problems where quantum could help, then let a small team, or even a single person, keep watch. 💡 Key themes ​Which parts of quantum are ready today, and how far to trust encryption that is in use but not yet proven​Why the first quantum promise failed, and what makes the second one more solid​How companies keep momentum and board support while progress stays slow​Whether quantum computing ends up with a few suppliers, the way AI did​Why Europe founds so many deep tech startups and struggles to grow them past Series B 👤 About the guestChristopher Trummer is MD Venture Lab Quantum / Semicon at TUM Venture Labs in Munich, where his team works with quantum and semiconductor startups from the research lab to their first investment. Having spent much of his career between science and business, he sees quantum from both sides: what the physics can already do, and when it starts to matter for a company that has to decide where to put its money. ⏱ Chapters ​00:00 Quantum beyond the computer​05:18 The algorithm that could break encryption​09:02 Why quantum won't have a ChatGPT moment​14:26 After the first promise fell through​24:31 Europe's missing growth capital​31:26 Drugs, fertilizer and superposition 🔗 Links ​Christopher Trummer on LinkedIn - https://www.linkedin.com/in/christopher-trummer-a58695108/​Eva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/​raidiant, Eva's responsible AI advisory - https://www.raidiant.eu/​TUM Venture Labs, Venture Lab Quantum / Semicon - https://www.tum-venture-labs.de/labs/quantum/​Global Deep Tech Report 2026, TU Munich - https://www.htgf.de/wp-content/uploads/2026/05/Global-Deep-Tech-Report-2026.pdf​NIST post-quantum cryptography standards - https://csrc.nist.gov/projects/post-quantum-cryptography​Related episode - Energy From Orbit: Can We Trust Power Beamed From Space? - Martin Soltau (EP 42) - [LINK: Eva adds]

    The Quantum Momentum: How Companies Stay Ready for a Computing Wave That Is Yet to Arrive - Christopher Trummer (EP 44)

  3. Oct 1

    How Close Can AI Get to the Human Mind? Persuasion, Trust and the Limits of Tech - Moran Cerf, Neuroscientist (EP 43)

    🎙️ Moran Cerf, Neuroscientist and Professor, Columbia Business School Moran Cerf, a neuroscientist and professor at Columbia Business School, studies what happens in our brains when we talk, decide and get persuaded, and lately what happens when the other side of the conversation is AI. His research shows AI can read people remarkably well: from recordings of speed dates, it predicted who liked each other better than human observers did. The same skill worries him: we raise our guard when a person tries to sell us something, but not when AI does it in much the same way. 🎧 Episode overviewMoran's starting point is neural: in a human conversation, brain waves tune to each other and a nod or a glance carries much of the meaning, while a question-and-answer exchange with AI holds about a fifth of that information. His verdict on AI's influence is blunt: "We're a lot simpler than we think we are," and purchase times, flight bookings and browsing history are enough for AI to infer personality and nudge even political views. Eva pushes on teams that give AI agents human names: helpful in a Ferrari pit stop, he says, and a hindrance in a boardroom, where too often "if AI says do this, we end up doing this." On trust, he draws a line between the many mundane choices AI already handles well and the rare ones that define us, like a partner or a profession, and warns that our trust spills from the first to the second. His reason for hope is practical: when the world or our life changes, the AI twin trained on our past fails, and we get called back in. 💡 Key themes Why AI can read and persuade people better than people can, and why we don't noticeWhen a human name for an AI agent helps a team, and when it distorts a decisionWhat the AI backlash has inherited from fifteen years of social mediaWhich choices to trust AI with, and how that trust spills over to the ones that define usWhether control handed over for convenience comes back when life changes 👤 About the guestMoran Cerf is a neuroscientist and professor at Columbia Business School whose work spans decision-making, persuasion and neural implants. He began by recording individual neurons to understand how the brain works, and several of his former PhD students now research AI at companies like Anthropic and OpenAI, which puts him right where brain science and AI meet. ⏱ Chapters 00:00 The feedback loop AI can't join05:46 AI reads attraction, then learns to persuade16:01 From neural implants to brain-like AI29:22 Trust it with groceries, not with a partner37:07 Giving up control, and the retwin42:29 In-between moments 🔗 Links Moran Cerf on LinkedIn - https://www.linkedin.com/in/moran-cerf-407187/Eva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/Columbia Business School - https://business.columbia.edu/staff/people/moran-cerfMoran Cerf's website - https://www.morancerf.com/Anthropic research Moran describes - https://www.anthropic.com/research/global-workspace Related episode - AI and The Cognitive Atrophy Trap: What Happens When We Let Tech Shortcut the Hard Parts of Learning - Tobias Burkhardt (EP 28) - https://open.spotify.com/episode/2vdo5QFoTroSQlZgEuRQin

    How Close Can AI Get to the Human Mind? Persuasion, Trust and the Limits of Tech - Moran Cerf, Neuroscientist (EP 43)

  4. Sep 24

    Energy From Orbit: Can We Trust Power Beamed From Space? - Martin Soltau, Space Solar (EP 42)

    🎙️ with Martin Soltau, Co-CEO, Space Solar Solar power collected in space and beamed down to Earth sounds like science fiction (it did to the space sector too, until reusable rockets arrived). Martin Soltau, Co-CEO of Space Solar, is building it in the UK, and he is confident about the technology: NASA demonstrated power beaming in the 1970s. The part of his timeline he can't engineer is trust: how quickly regulators and the public accept energy that comes down from orbit. 🎧 Episode overviewIt starts with Martin's verdict on Europe's energy transition: the UK and Germany, he says, have "tested it to destruction", and households pay for it with expensive electricity and weaker security. His case for orbit is concrete: a panel in space produces eight to 13 times what it would in the sunniest place on Earth, robots assemble the satellites from coffee-table-sized modules, and the beam can be directed wherever power is needed. Eva pushes on the catch: if launch still runs through a handful of mostly American providers, does Europe swap one dependency for another? On trust, Martin points to fracking (a technology that worked and still lost the public) as his warning, and explains why he answers the safety question with wifi frequencies and intensities anyone can measure. 💡 Key themes Why a grid built on free sunshine and free wind still delivers expensive electricityWhether public trust can be earned before the first satellite flies, or only afterwardsWhat wireless power changes about energy security when undersea cables are already being targetedWhether governments that fund this will accept standards letting a rival's satellite use their receiver 👤 About the guestMartin Soltau co-founded Space Solar in 2022 to build solar power satellites that deliver continuous electricity to receivers on the ground. He came from inside the space sector, where the idea sat on the science fiction pile until SpaceX made reusable launch commercial. Today his work is as much about convincing governments and regulators as about engineering. ⏱ Chapters Tested to destruction (starts with: "I think it's the big challenging problem for humanity ...")Why orbit beats the sunniest place on Earth (starts with: "So I suppose it starts with the fact that if you put a solar panel into space ...")Launch, and who controls it (starts with: "Yeah, that's a very good question. I mean, if we look at wind and solar and batteries ...")Standards between rivals (starts with: "Yes, I think there are many examples of where nations have collaborated ...")Coffee-table modules and crawling robots (starts with: "Yes. Well, if you came to space solar at Harwell ...")The trust question (starts with: "I think you use a really interesting word there, trust ...") 🔗 Links Martin Soltau on LinkedIn - https://www.linkedin.com/in/martin-soltau-44596619/Eva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/Space Solar - https://www.spacesolar.co.ukInternational Telecommunication Union - https://www.itu.intRelated episode - Sten Saluveer on what decides AI trust in Europe - https://open.spotify.com/episode/0gPKehnoNppMY0y3Jvdry0

    Energy From Orbit: Can We Trust Power Beamed From Space? - Martin Soltau, Space Solar (EP 42)

  5. Sep 17

    Entering the Era of AI Power: An Estonian View on Europe's Way - Sten Saluveer, Telia (EP 41)

    🎙️ Sten Saluveer, Growth & Innovation, Telia What decides AI trust in Europe - the size of the model, or whether people recognise themselves in the society using it? Sten Saluveer helped write Estonia's digital state agenda and now builds at Telia, and he argues Europe enters the era of AI power holding more than it thinks: a working example, and the diversity to invent its own category. For leaders deciding where their organisations stand in that game, this is a view from a country that stopped writing roadmaps and started shipping. 🎧 Episode overview Sten and Eva move from why a builder who ran innovation at Cannes chose an Estonian telco, to what an agent-driven state actually looks like when more than a hundred and twenty agentic services already run and the plan reads like scenes from a citizen's life. They sit with the uncomfortable side: a quarter in which Estonians lost around thirty million euros to cybercrime, and a security goal that paradoxically aims to keep satisfaction exactly where it is. From Cannes, Sten brings the generational fault line - leaders over forty who see adaptation as given, talents under thirty who experience AI as a threat to their work and their environment - and the question of whether Europe can turn its diversity into its own category rather than playing everyone else's game. 💡 Key themes Whether a state can make itself agent-driven and keep the citizen in controlWhy holding cybersecurity satisfaction at today's level counts as ambitionWhat Europe's own category could be when it stops measuring itself in gigawattsThe generational split on AI that surfaced at Cannes, and what it demands of leadersWhether a telco can become the trusted digital partner of the AI economy 👤 About the guest Sten Saluveer is an Estonian innovation strategist who has spent two decades translating between the market, the technology, and the governance layer - from heading Cannes Next at the Marché du Film to advising the Tallinn Digital Summit. He contributed to Estonia's digital state agenda, where he insisted the plan contain scenarios from ordinary life, and he recently joined Telia heading their growth and innovation initiatives. What makes his perspective matter here is the vantage point: he argues Europe's case from inside a small country where the agent-driven state is already operational. ⏱️ Chapters [00:00] Why a builder chooses a telco when the era turns[11:41] An action plan, not a roadmap - the state that ships[18:11] Standing still as ambition - the cybercrime paradox[23:15] When infrastructure has to assume it is a target[28:03] Cannes: the generation that experiences AI as a threat[38:26] Reclaiming agency - Europe's own category 🔗 Links Sten Saluveer on LinkedIn - https://www.linkedin.com/in/stensaluveer/Eva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/Telia - https://www.telia.ee/eraEstonia's digital state agenda - https://digital-skills-jobs.europa.eu/en/initiatives/national-strategies/estonia-estonian-digital-agenda-2030Related episode - 'Tech and Democracy with Nexus Politics' https://open.spotify.com/episode/3WVnBsaz3LfFPwn6xsX1N6

    Entering the Era of AI Power: An Estonian View on Europe's Way - Sten Saluveer, Telia (EP 41)

  6. Sep 10

    Captured or AI Created? Tech Ethics of Photography - Noah Wetzel (EP 40)

    🎙️ Noah Wetzel, Photographer What happens to AI trust when the most real photographs are the ones people refuse to believe? Noah Wetzel spent nine months and 600 hours preparing to capture 84 seconds of a total solar eclipse - then had to prove the result came out of a camera. This one is for anyone whose genuine work now meets default suspicion, and anyone deciding where AI belongs in a craft that has to stay real. 🎧 Episode overview Noah captured a wingsuit athlete flying through totality in Spain: flashes triggered from a mile away, a subject moving at 150 miles per hour, two frames per athlete, no second takes. The conversation follows what came after the shutter - viewers at his own shows assuming the work is fake, magazine editors auditing every edit in his raw files, and a disclaimer on Instagram stating his photos are camera-captured. He is blunt about his own AI use: hundreds of hours on ChatGPT planning the shot, AI saving him days in processing, and a hard line at the frame itself. Where that line sits, and whether it survives cameras that can generate content on demand, stays open until the end. 🔍 Key themes Why extraordinary authentic images now read as fake by default, and who ends up carrying the burden of proofWhere AI strengthens a craft and where it disqualifies the result - the captured versus created distinction Noah refuses to blurWhat "the ethics of photography" means when every edit is contestable and everyone draws the line differentlyWhat happens to that line when the camera itself can add what was never thereHow you perform inside 84 seconds when preparation is the only thing left to trust 📸 About the guest Noah Wetzel is an adventure and action-sports photographer, and the person behind what may be the most complex photograph ever attempted: a Red Bull Air Force wingsuit athlete crossing a total solar eclipse. The detail that matters for this episode is what he does after the work is done - he attaches a statement to his published images confirming they were captured by a camera, because the audience's default assumption has flipped. He argues the case for craftsmanship from inside a working life that uses AI daily. ⏱ Chapters [00:00] 600 hours for 84 seconds[04:33] Trusting preparation when nothing can be paused[16:30] What generative AI changes about believing a photograph[24:02] Why real work now carries a disclaimer[31:41] Where the in-camera line starts to dissolve 🔗 Links Noah Wetzel's website / portfolio — ⁠www.wetzelgallery.com⁠Noah Wetzel on Instagram, ⁠www.instagram.com/noahwetzel⁠Noah Wetzel on Facebook: ⁠www.facebook.com/noahwetzelphoto⁠[LINK: Related episode — suggest a Pillar 1 episode on trust in AI systems, Eva to match to episode number]Eva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/Documentary Released on PBS Recently - April 8th 2024 Total Solar Eclipse in Hill Country Texas. The film has received 17 film festival awards and been aired 1700+ times in the last month across 39 different states in the United States. You can watch the documentary online here:https://www.pbs.org/show/eclipse-outlaw-chase/

    Captured or AI Created? Tech Ethics of Photography - Noah Wetzel (EP 40)

  7. Sep 3

    AI and Sustainability: Lessons from Both Transformations - Dr. Kiri Trier (EP 39)

    🎙️ Dr. Kiri Trier, Chief Marketing & Growth Officer at Risk Management Partners, part of Munich Re 👋🏻 Opening paragraph Every C-level claimed sustainability expertise overnight. Dr. Kiri Trier watched it happen, and she sees AI trust heading the same way. She spent more than a decade leading sustainability transformation before the current wave, and now runs growth inside Munich Re's climate risk business while the second hype cycle unfolds around her. This one is for leaders who suspect the pace of AI adoption has outrun the understanding beneath it. 🧭 Episode overview In this episode, Dr. Kiri Trier maps the sustainability cycle onto the AI one: the expectation management problem when nobody knows the consequences, the human resource comparison that sustainability never triggered, and why she laughed the first time she heard "human in the loop". From inside reinsurance she raises the question with no settled answer yet - when a model judges wrong, what about the liability? And she closes with a definition that reorders most AI governance thinking: trust arrives only as an outcome of credibility and reliability. 💡 Key themes When a model judges a policy wrong, who carries the liability - and why the autonomy still lies on the human sideWhether the EU AI Act repeats the German supply chain law: does checklist regulation produce compliance or trust?What greenwashing predicts about AI washing, from someone who watched the first one formCan trust be installed at all, or only earned as an outcome of credibility and reliability?The gut feeling as a leadership instrument the LLM does not have 👤 About the guest Dr. Kiri Trier is Chief Marketing & Growth Officer at Risk Management Partners, a Munich Re company localising climate risk exposure. The detail that matters for this episode: seven years before the current wave, she published on how AI would change sustainability management - which means she entered the AI transformation already knowing how a hype cycle ends. ⏱️ Chapters [01:31] Trust is a behaviour - so what is trust in a technology?[04:09] The overnight experts return[09:32] Human in the loop, turned around[16:10] The foresight existed - Sustain what?[26:00] When regulation becomes Excel lists 🔗 Links Dr. Kiri Trier on LinkedIn - https://www.linkedin.com/in/dr-kiri-trier-97bb1611/Eva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/Risk Management Partners website - https://www.munichre.com/rmp/en.htmlRelated episode: 'Designing regenerative futures' with Karel Golta - https://podcasts.apple.com/de/podcast/the-in-between-tech-and-trust-podcast/id1828521905?l=en-GB&i=1000721908996

    AI and Sustainability: Lessons from Both Transformations - Dr. Kiri Trier (EP 39)

  8. Aug 27

    Building Trustworthy Open Source AI - Sebnem Erener (EP 38)

    🎙️ with Sebnem Erener, AI Governance & Policy Expert 🎧 Opening Europe is investing billions in open source AI as the trustworthy path, and this episode asks a simple question about that plan: does open actually mean trustworthy? Sebnem Erener works on AI governance, and her answer is no, or at least not on its own. Open source spreads access to a model. Whether you can trust it, and who answers when it goes wrong, are different questions with different answers. 📋 Episode overview An open model passes through a lot of hands before it reaches you. Sebnem counts five or six: uploaded to a platform, hosted somewhere else, routed, fine-tuned, served. The law usually wants one company to hold responsible, and that's not how the open ecosystem works. She spent years at Klarna as its Head of AI Legal, Regulatory and Ethics, and she now studies where this gap sits. We talk about why she cares more about trustworthiness than trust, what open source can and can't fix on its own, and why she thinks a person can be held liable but a model never can, no matter how independently it acts. It's a conversation for anyone building on an open model, or writing the rules for one, who wants to know who's actually accountable. 🧭 What you'll hear Why an open model can leave you less sure who's liable than a closed oneThe line between real openness and what Sebnem calls "open washing"Where Europe should be patient, where it should be bold, and why one timeline for all of it is a mistakeWhy calling AI "agentic" doesn't move the responsibility onto the machineThe thing Europe is short on, which isn't innovation 👤 About Sebnem Erener Sebnem Erener works on frontier AI governance and policy across EU, UK, and US frameworks. Before going independent she was Head of AI Legal, Regulatory and Ethics at Klarna, so she has handled these questions inside a company and not only in theory. She's an advocate for open source AI, and she's precise about what it does and doesn't solve. Her forthcoming paper on the governance of open source AI intermediaries runs underneath a lot of this conversation. ⏱️ Chapters [00:00] What we mean when we say "trust"[05:00] The five or six hands on every open model[09:00] When open makes accountability harder[12:00] One package, four timelines: patience and boldness for Europe[14:30] Why "agentic" lets no one off the hook 🔗 Links Sebnem Erener on LinkedIn - https://www.linkedin.com/in/sebnemerener/Sebnem Erener website - https://www.sebnemerener.comEva Simone Lihotzky on LinkedIn - https://www.linkedin.com/in/evalihotzky/European Commission tech sovereignty package - https://digital-strategy.ec.europa.eu/en/policies/eu-tech-sovereigntyOpenEuroLLM - https://openeurollm.euRelated episode - beyond the black box with Prof Dr Tina Weisser: https://open.spotify.com/episode/6yKMLQNnDoa6NqqClJrryN?si=3AWXxowURVujJL5m3zD5QQ

    Building Trustworthy Open Source AI - Sebnem Erener (EP 38)
See All (45)

About

How do we build, break, and rebuild trust in a world shaped by technologies? The In-Between Tech & Trust is a weekly conversation about the human side of tech. Host Eva Simone Lihotzky, founder of raidiant (responsible AI advisory), talks with guests from business, politics, neuroscience, and tech about what they have learned about trust, and what they would pass on to you. Eva was formerly co-founder and MD of the Serviceplan AI Lab, holds the Global Moral Chair at the Value AI Institute and co-authored "10 Moral Questions: How to Design Tech & AI Responsibly".

Information

You Might Also Like