Advertise on podcast: AI Snacks With Romy & Roby

SummaryPart 2 builds on the previous episode, describing how to establish AI governance in an organization, and what AI literacy truly means for business results. Anthony Scriffignano, PhD is one of the world's best data scientists — a thinker who has spent more than 40 years at the intersection of linguistics, advanced algorithms, business intelligence, and the very human problem of making good decisions from imperfect information. Anthony is currently a Distinguished Fellow and member of the Alfred Lee Loomis Innovation Council at The Stimson Center. He also serves on advisory committees in government, the private sector, and academia, and has briefed the National Security Telecommunications Advisory Committee and contributed to reports to the President on Big Data Analytics and Emerging Technologies. He served on the OECD Network of Experts group on implementing Trustworthy AI. His most recent publication is a co-authored chapter — "Illuminating the Invisible: AI's Role in Understanding Complex, Opaque, and Dynamic Ecosystems" — in the Springer volume Artificial Intelligence Revolution (2026, edited by Max Rangeley and Nicholas Fairfax), co-written with Dr. David Bray. Key Takeaways: Compliance is what you must do; governance is what you should do. Governance costs money, time, and opportunity. The right question is not "how do we maximise governance?" but "where do we sit in the continuum between maximum safety and maximum agility, and are we there on purpose?"  The human-in-the-loop section challenges the common assumption that more humans equals better decisions. Anthony identifies three limits: humans cannot always make decisions fast enough; even the same human given the same decision repeatedly will not always make the same call; and humans have bias, fatigue, and optimism. The emerging solution is cognitive AI that watches experts making decisions, converges on patterns, and eventually can substitute a synthetic human into the loop for low-stakes votes. Agentic AI is described as a system of semi-autonomous entities, each doing a small piece of a problem, orchestrated together, some inside the enterprise and some outside. Auditing every individual agent is not feasible. The answer is a clinical layer that monitors the health of the agentic ecosystem as a whole — taking the pulse, measuring blood pressure — and raises flags when the system-level behaviour deviates from its guardrails. The language to do this does not fully exist yet. On shadow AI: Anastassia's survey found 64% of respondents in SMEa are using AI tools their employer has not sanctioned. Anthony suspects the real number is higher and points to non-response bias. His prescription is not HR enforcement, but understanding the underserved need: if employees are using shadow AI, the organisation has failed to provide them with the tools they need. Create a safe sandbox, serve the need, and watch what they build. Chinese LLM approach takes a fundamentally different approach from the US— leaner, faster, cheaper, focused rather than voracious. AI literacy versus AI fluency. Fluency — knowing the buzzwords, knowing how to prompt — is the lowest form of useful knowledge. Literacy means knowing what you do not know: understanding the limits of the tool, the provenance of the output, and when you should not trust it.  Hyperlinks:Anthony’s LinkedIn Stimson Center Fellow Profile  https://www.stimson.org/ppl/anthony-scriffignano/ Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)  AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book

SummaryThe conversation focuses on how to build a data strategy addressing the needs of a business first, not what is available on the market in terms of an AI model. Anthony Scriffignano, PhD is one of the world's best data scientists — a thinker who has spent more than 40 years at the intersection of linguistics, advanced algorithms, business intelligence, and the very human problem of making good decisions from imperfect information. Anthony is currently a Distinguished Fellow and member of the Alfred Lee Loomis Innovation Council at The Stimson Center. He also serves on advisory committees in government, the private sector, and academia, and has briefed the National Security Telecommunications Advisory Committee and contributed to reports to the President on Big Data Analytics and Emerging Technologies. He served on the OECD Network of Experts group on implementing Trustworthy AI. His most recent publication is a co-authored chapter — "Illuminating the Invisible: AI's Role in Understanding Complex, Opaque, and Dynamic Ecosystems" — in the Springer volume Artificial Intelligence Revolution (2026, edited by Max Rangeley and Nicholas Fairfax), co-written with Dr. David Bray. Key Takeaways:Data before models. Most AI failures begin in the data, not in the model. Ask the three Ps before you touch a model: Provenance, Permissible use, Perturbation. Decision elasticity. The right question is not "is this data perfect?" but "how wrong could it be and would I still make the same call?" If you cannot answer that, you are not ready to decide. The walled garden method. You can detect disruption inside a system you cannot observe directly by modelling the ecosystem immediately outside it — the haversine annulus, Z-scores, Jensen-Shannon divergence. Start with needs, not tools. Four-box needs analysis: known met, known unmet, unknown met (the danger box), unknown unmet (the blue sky box Synthetic data compounds error. Pre-LLM synthetic data let you estimate error. Post-LLM, models are generating training data that other models consume, and error compounds multiplicatively rather than averaging out. Hyperlinks:Anthony’s LinkedIn Stimson Center Fellow Profile  https://www.stimson.org/ppl/anthony-scriffignano/ Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)  AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book Chapters 00:00 Introduction to the importance of data in AI 02:03 Meet Dr. Anthony Scrifignano and his work 03:21 The distinction between having data and useful data 05:21 Understanding decision elasticity in AI models 07:42 Messiness of real-world data and ecosystems 10:01 Detecting ecosystem disruptions without internal data 15:03 Fraud detection and anomaly identification in disasters 19:42 Starting with needs analysis for data strategy 23:05 The importance of first principles and understanding business needs 26:42 The pitfalls of applying rigid methodologies without understanding 30:10 Synthetic data, its challenges, and generative AI risks 42:58 The core of AI is about foundations, not architecture

In episode 95 of AI Snacks with Romy and Roby, Dr. Anastassia Lauterbach sits down with science fiction author Peter Cawdron — 37 books, NASA and SpaceX consultations, and a First Contact series 36 novels deep — to explore what realistic science actually tells us about living on Mars, what AI can and cannot do, and what it would take for a machine to develop genuine values. The conversation moves from the brutal physics of Martian habitation to the Chinese Room thought experiment, from Maslow's hierarchy of needs to the moral arc of the universe. A rare episode where science fiction, engineering, and philosophy arrive at the same question: what makes us human, and can we build that into a machine? Peter Cawdron is a science fiction author based on the Sunshine Coast, Queensland, Australia, originally from Auckland, New Zealand. He is the author of 37 books, known for grounding his fiction in the hardest available science — consulting with NASA flight controllers, SpaceX engineers, and space mission specialists to ensure his stories adhere as closely as possible to physical reality. Key Topics The influence of real science on sci-fi storytelling and how experts shape the research processLiving on Mars — radiation, soil toxicity, underground habitats, and the engineering of survivalThe Chinese Room thought experiment and why LLMs are not reasoningThe Gutenberg Press phase of AI — transformational but still to be determinedJevons' Paradox and why AI may create more jobs than it eliminatesArtificial intelligence vs. artificial consciousness — a fundamental distinctionMaslow's hierarchy of needs and why AI is stuck at the baseThe Warren Buffett problem — why digital avatars are not the personA proposed test for AGI — give it all human knowledge up to 1900 and ask it to find EinsteinHow to give AI values, morals, and culture — and why it matters for avoiding a Terminator futureThe seesaw between selfishness and cooperation that defines human civilization Chapters 00:00 Introduction to the episode and guest 01:01 Peter Cawdron's background and journey into sci-fi 02:17 The role of science and curiosity in human progress 04:07 Historical perspective on science and its impact on society 05:31 How real science influences sci-fi writing 08:02 Researching realistic space habitats and challenges on Mars 10:30 The plausibility of colonizing Mars and environmental challenges 17:36 AI as an existential threat and the concept of dumb AI 19:18 The exponential growth of AI and its societal implications 21:33 AI as a Gutenberg Press revolution and its future impact 25:14 The importance of values, morals, and culture in AI development 30:28 The Chinese Room thought experiment and AI understanding 35:42 Characters and themes of digital avatars and grief 41:30 Consciousness, self-awareness, and the brain's complexity 44:06 AI morality, values, and the risk of Terminator-like futures 48:47 Human traits — loyalty, trust, and the essence of humanity 53:45 Upcoming projects and books by Peter Cawdron Guest Links Peter Cawdron on AmazonFirst Contact seriesTo Live and Die on Mars (August 2026)To Live and Die on Earth (September 2026)The Song of the Sirens (upcoming) Follow AI Snacks with Romy and Roby Podcast: romyandroby.aiInstagram: @romyandrobyLinkedIn: Dr. Anastassia LauterbachBook: Romy, Roby and the Secrets of Sleep — available on Amazon

Summary Anastassia and Anne Cantera discuss voice interfaces and key design principles for making voice chatbots "right." Anne designs voice experiences at scale, builds orchestration tooling for multi-agent systems, and teaches what she calls omnimodal design. Anne created VoiceofAI.io - a free AI upskilling resource for designers. Key Takeaways: Getting good AI output is a skill: provide requirements, design systems, and data so the model never has to make things up. Never let AI write for you - let it interview you. That's the difference between AI-assisted and AI-generated — a thought innovator versus a slop generator. The future is omnimodal design - technology that offers what you need, in the moment you need it. Disclosure is non-negotiable. Voice AI can now listen and speak at the same time - making it easier than ever to fool people. Pretending a bot is human is a dark pattern. Trust is gained in drops and lost in buckets. Companies that ship half-baked voice agents accumulate “emotional debt.” Fast, easy, cheap: Voice projects are none of the three. Replacing humans is backfiring. From flopped plans to outsource support to agents, to companies quietly rehiring. Narrow domains win. A small, women-owned bank with one product can build a better chatbot than an enterprise with 110 products. Bring in experts early — and learn to spot them. Leadership ego (“we don't need consultants”) makes messes that cost more to clean up later. But AI literacy comes first: an influencer with 500,000 followers and no AI work history is not an expert. LLMs are a step, not the final frontier. Chapters: 00:00 Intro: when the machines around us began to speak 02:24 Welcome Anne: dialing in from Long Island 03:26 From chef to conversation designer: an unusual path into AI 05:52 Self-taught in NLP: free chatbots, LexisNexis, and skills no school teaches 09:27 Omnimodal design: the perfect modality at the perfect moment 11:46 Being a good AI operator: spec-driven work and the interview technique 14:07 Human skills come before prompts 16:34 Voice tech today: machines that listen and speak at once 18:55 Ethics and dark patterns: why Anne left corporate 21:16 Emotional debt: trust is gained in drops and lost in buckets 24:42 Elastic organizations: core beliefs before AI design 26:59 Theory of mind: why machines cannot relate to us 31:39 Enterprise vs. everyone else: why small companies build better bots 36:42 Domains and jobs: machine translation, where nobody was laid off 39:42 Leadership ego: bring in your experts early 40:37 AI literacy first: wannabes, influencers, and real experts 43:49 The upskilling epidemic — and why VoiceofAI.io is free 49:20 Beyond transformers: LLMs are a step, not the final frontier 52:43 Robotics: industry first, our homes later 54:13 The AI class divide: haves, have-nots, and AI teachers 55:29 Closing: two more episodes promised — design and ethics   Hyperlinks: Anne Cantera LinkedIn Elementyl Intelligence Website The Voice of Intelligence Design Handbook and Website Anne Cantera Article on Conversational Systems Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)  AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book

Summary: Can a machine put itself in your shoes? In this episode, Dr. Anastassia Lauterbach talks with Andrew Phipps - a PhD researcher in philosophy at the University of Sussex focusing on AI ethics and empathy, and host of the Empathy Unbound podcast, whose 120+ conversations with researchers, clinicians, founders, psychologists, and neuroscientists serve as his philosophical field work. Andrew left a corporate career - managing over a thousand people across fifty countries — to research the question he couldn't find answered anywhere: what does a positive relationship between humans and AI actually look like? Key Takeaways: Agency is the heart of AI literacy. You can't decide about something you don't truly understand Young people are not one group. Some refuse AI on environmental grounds, some fear the graduate jobs their parents had are vanishing, some see an entrepreneurial opening, and some embrace it for everything. Educate the educators. “Don't use AI for your homework” is not AI education. Teachers are expected to guide students without being trained themselves. Licenses are not a strategy. Ten thousand AI licenses without training is a pen without writing lessons. Nobody knows who owns AI. Unlike ordinary IT, AI belongs to the whole business — which is why it often belongs to no one. The average Chief AI Officer lasts eighteen months: barely time to learn the organization, let alone change it. AI itself cannot be ethical. It answers based on how you structure the question and what it predicts comes next. Ethics and morality live with the individuals and companies that set standards for its use. Readiness before rollout. Roughly 30% of people admit to sabotaging AI programs they fear. The trust gap is about power, not answers. Around 70% of people in China express trust in AI versus roughly 30% in the US — where trust in AI now rivals trust in politicians. What people distrust is not the output but AI's position in society: enormous power concentrated in a few largely unregulated companies led by a handful of charismatic founders. Purpose is the hidden stake. If AI and universal basic income ever detach humans from the need to work, psychologists fear for the human psyche: we have always operated with purpose. AI cannot feel empathy — and pretending otherwise is dangerous. Empathy means understanding another's feelings and acting as a consequence. AI predicts; it does not feel, reason, or possess theory of mind Chapters: 00:00 Introduction to AI Literacy and Human Agency 03:01 Andrew Phipps: Transitioning to AI Ethics Research 06:13 Understanding Human Agency in the Age of AI 07:47 Generational Perspectives on AI and Employment 11:51 The Role of Education in AI Literacy 13:56 Preparing Educational Systems for AI Integration 18:30 Corporate Approaches to AI Integration 22:51 Ethical Considerations in AI Development 28:27 Trust in AI and Its Societal Implications 31:16 Purpose and Employment in an AI-Driven World 33:55 Empathy and AI: Can Machines Understand Us? Hyperlinks: Andrew Phipps on LinkedIn Empathy Unbound ChannelAnastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)  AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book

Summary: Anastassia sits down with Yaron Levi, CISO of Dolby Laboratories and one of the world's most respected security leaders, to unpack the reality of AI-powered cyberattacks: from AI-generated phishing at massive scale to the first documented AI-orchestrated cyber espionage campaign. Yaron explains why the fundamentals of security haven't changed — but the speed and volume of attacks have — and why decades of technical debt are now every organization's soft underbelly. Yaron has over 20 years of hands-on experience across enterprise security, startups, and the intersection of AI and education. Prior to Dolby, Yaron was CISO of Blue Cross and Blue Shield of Kansas City, Deputy CISO at Cerner Corporation, an Information Security Business Partner at Intuit, and a Security Architect and Product Manager at eBay. He is a Research Fellow of the Cloud Security Alliance (CSA), a graduate of the FBI CISO Academy, a Boardroom Certified Qualified Technology Expert (QTE), and a venture advisor to multiple VCs and security startups. Key Takeaways: AI democratizes attack capability with LLMs lowering the bar; The fundamentals haven't changed: Breach causes remain the same year after year (compromised credentials, misconfigurations, phishing). AI simply finds and exploits them dramatically faster; Technical debt is the new attack surface with forgotten accounts, stale permissions, and abandoned servers; Third-party risk needs a rethink. Questionnaires and certifications don't guarantee anything; measuring how much you can trust a partner matters more than measuring what they might be hiding; Cybersecurity is still a young industry without one common standard. Despite ISO, NIST, and many frameworks, CISO collaboration remains largely informal — community and information sharing are essential; Start with the business, then threat model and communicate risk with a "pyramid of security needs;" Breaches rarely kill big brands — but can kill small companies. Reputation matters, yet resilience and the ability to recover matter more; repeated, unlearned-from incidents are what truly erode trust; Advice to vendors: be partners, not sellers; Data quality is king. Garbage in, garbage out: AI forces organizations to audit what data they have, retire stale data, and sometimes give old data new life; Optimism for the next generation. Like the cloud wave of 2009–2010, AI will create opportunities we can't yet imagine — it has never been easier to learn, build, and start a company. Chapters: 00:00 Introduction to AI and Cybersecurity Challenges 03:02 The Dual Nature of AI: Opportunities and Threats 05:15 Evolving Cybersecurity Landscape: The Role of AI 10:23 Addressing Speed and Human Factors in Cybersecurity 13:05 Third-Party Risks in the Age of AI 18:38 Collaboration Among CISOs: A Collective Approach to Security 22:45 Communicating Cybersecurity Risks to Leadership 24:02 Understanding Cybersecurity Needs 26:56 The Pyramid of Security Needs 27:54 The Role of Trust in Cybersecurity 30:29 Advice for Cybersecurity Vendors 34:29 Surprising Trends in Cybersecurity 37:30 The Impact of AI on Data Analysis 40:30 Navigating the Job Market in the Age of AI Hyperlinks: LinkedIn Yaron Levi X / Twitter Yaron Levi Dark Reading author page (articles) https://www.darkreading.com/author/yaron-levi Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)  AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book

Summary: What does knowledge mean in the age of AI? Anastassia talks to César A. Hidalgo, known for pioneering work on economic complexity, data visualization, and applied artificial intelligence. Key Takeaways: AI's knowledge is jagged. Think of the economy as a game of Scrabble. AI hands you many Rs, As, and Ts but no Us or Es. Work that needs the letters AI provides accelerates dramatically; work that depends on the missing ones becomes the bottleneck. Someone with real architecture experience gets far more out of a coding agent than someone building blind. Complementarity beats substitution. An organization's AI results depend on whether its people hold skills complementary to what the model does well. If they do, expect enormous acceleration; if they don't, expect half-baked output. Judgment is the scarce human asset. AI is very smart but does not choose directions. The future belongs to organizations built around people who can see where a system is heading and form teams around that insight. Judgment is hard to observe, but it is essential in a changing, uncertain environment. Progress does not require understanding. For most of history, technology advanced through practices that worked without anyone knowing why. Even today, many discoveries come from an unexpected impurity or an exploratory detour rather than a fully prescriptive theory. Exploration remains central. Use AI as a telescope, in your own field. AI is a tool for exploring codified knowledge, but exploration is always tied to the skill of the explorer. Plausible-sounding errors are much easier to catch in a domain where you already have expertise. Knowledge loves density. Ecuador spent one percent of GDP on Yachay, a greenfield “city of knowledge” that wanted to be good at everything in ten years; construction stalled, the supercomputer is unmaintained, and water still arrives by truck. Beijing's Zhongguancun instead bet on a single street, with government-private guiding funds, and became the template for China's shift from mass manufacturing to mass innovation. Most AI pilots fail because that is how evolution works. Attention and money are flooding into AI, so many attempts fail and a few succeed and get copied, just as web companies did in 1997. Chapters: 00:00     Intro / Exploring Knowledge in the Age of AI 02:26     Cesar Hidalgo: A Journey from Physics to Economics 04:49     Understanding Knowledge: Factual, Conceptual, and Procedural 07:40     The Laws of Knowledge and Economic Growth 09:02     Tacit Knowledge: The Hidden Driver of Innovation 14:01     Capturing Tacit Knowledge in Organizations 19:08     The Decay of Knowledge and Its Implications 20:25     AI and Human Learning: A Complementary Relationship 24:45     Digital Twins: The Future of Knowledge Representation 28:32     Understanding vs. Knowledge: The Key to Innovation 29:06     The Role of Understanding in Technological Progress 30:31     Exploration and AI: Navigating Knowledge 32:34     The Value of Knowledge in Corporate Acquisitions 34:33     Innovation Ecosystems: Lessons from Success and Failure 38:10     Advising Fortune 50 Companies on Knowledge Management 44:46     Leadership and Decision-Making in Innovation 49:27     Challenges of AI Implementation and the Role of Scarcity Books by César Hidalgo: The Infinite Alphabet: And the Laws of Knowledge (Penguin / Allen Lane, 2025): Penguin UK Why Information Grows: The Evolution of Order, from Atoms to Economies (Basic Books, 2015) How Humans Judge Machines (MIT Press, 2021) The Atlas of Economic Complexity: Mapping Paths to Prosperity (MIT Press, 2014) Hyperlinks: César Hidalgo Website Center for Collective Learning Toulouse School of Economics César Hidalgo profile Company Datawheel César Hidalgo X (Twitter) César Hidalgo LinkedIn Anastassia Interviewing César Hidalgo Anastassia Lauterbach - LinkedIn AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book

Summary: Anastassia sits down with Chetan Sharma, one of the most respected strategists in global wireless. For 25 years, Chetan has been an independent strategist advising CEOs and CTOs of the world’s leading wireless companies, e.g., NTT DoCoMo, AT&T Wireless, China Mobile, Qualcomm, Samsung, and Sprint, and he has served on the advisory boards of Ericsson, Telefónica, Kymeta, NextNav, and numerous startups. He wrote Mobile Advertising: Supercharge Your Brand in the Exploding Wireless Market and Wireless Broadband: Conflict and Convergence — with titles adopted in corporate training programs and university courses at NYU, Stanford, and the University of Tokyo. He is the editor of the Mobile Future Forward book series. Key Takeaways: Every wireless wave follows the same curve. Voice, messaging, and data each peaked once penetration hit 80–90%, then revenue declined. Operators must invest in the “fourth wave” — platforms, applications, and enterprise services — before the data curve turns down. China proved the fourth-wave thesis. By transforming from pure comms players into IT players, Chinese operators generated close to $100 billion in enterprise and platform services over five years — a scale the rest of the world hasn’t matched. AI’s home in the network is the platform layer. Intelligence will live everywhere — in the RAN (Radio Access Network), at the edge, on devices — but the platform layer orchestrates it all: energy, RAN management, compute, and the “intelligence traffic” of tokens across billions of nodes. Compute and comms are finally converging, enabling more sophisticated algorithms for spectral efficiency and network manageability. AI already delivers hard savings. Intelligent on/off orchestration of cell-site carriers alone can yield 40–60% energy savings, and AI-driven operations can resolve issues before humans even see the ticket. Security is a multi-level battle. AI lets attackers automate vulnerability hunting across operator, enterprise, and national-security layers — including legacy systems. Talent is strategy. Operators outsourced R&D and product development for decades; AI is the chance to bring it back in-house. New ventures need freedom to run. AT&T’s connected-devices unit succeeded because it operated as an independent entity with full support from the top. Europe is falling behind. A steep patent decline since ~2008–09, a weak LTE and 5G cycle, and long lags between intention, budget, and execution — compounded by AI Act–driven startup exits — put the EU at risk of missing the AI cycle too. The best telco CEOs are storytellers and portfolio managers. Hold multiple truths at once. Borrowing from the Buddhist Catuṣkoṭi: AI is simultaneously a bubble, transformative, and disruptive. Don’t wed yourself to one truth — let the situation guide you. Chapters: 03:13 From 1G fraud detection with neural networks to 25 years of strategy consulting 06:09 The waves of wireless: voice, messaging, data and beyond 12:16 Where AI lives in the network stack — and the new security battlefield 17:01 AI-run networks: What does it mean? 21:10 Disaggregating the RAN, exposing network intelligence, and reinventing the operator 31:17 CAPEX, economics of compute, sovereign AI and why Europe is behind 39:45 What makes a great telco CEO Hyperlinks: Chetan Sharma Consulting Chetan Sharma LinkedIn Chetan Sharma X (Twitter) Mobile Future Forward (executive summit & book series) https://www.mobilefutureforward.com Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)  AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book

Summary: Josh Gould, Group CEO of TheBigWord – one of the world's largest language services providers, operating in over 80 countries with a workforce of 15,000 – explains why today's LLMs look more human than 2013's systems without actually being more accurate, how his company's orchestration layer routes each job to the model that is best at that language pair and domain, and why human linguists have shifted from translating to tuning, reviewing and editing while translation demand doubles every year. Joshua (Josh) Gould has spent his career at the intersection of language technology and human expertise – including TheBigWord's multi-year collaboration with Google on human feedback for neural machine translation. TheBigWord provides translation and interpreting across borders, hospitals, courts, and healthcare systems, giving people language access when they need it most. Beyond language technology, Josh is a co-founder, investor, and board member at a healthy frozen plant-based food company, and a frequent podcast guest on AI, entrepreneurship, and the future of work. Key Takeaways: Models are old; scale is new. The architectures behind today's AI existed on paper for decades – chips, electricity, and internet-scale data made them real. The baseline of every model is still human. LLMs didn't beat 2013. Current models don't score meaningfully better than the best neural Machine Translation of a decade ago – they look more human, while filling gaps with estimations that can make them slightly less accurate. Orchestration is the moat. No single model is best at everything; TheBigWord's layer routes French–Dutch orthopedic content to one engine and English–Russian general content to another, tuned continuously by client and linguist feedback – pushing accuracy from ~94% toward 98–99%. The linguists didn't disappear – their jobs changed. Translated content demand doubles every year; unit rates fell, volumes exploded, and translators became editors, reviewers and tuners. The predicted extinction never happened. Tokens are the hidden bill. Interpreting (voice-to-voice) consumes six to ten times the tokens of written translation, and today's AI prices are loss-making subsidies: expect roughly a doubling once providers must be profitable – machine interpreting at 75% of human cost today could be 150% tomorrow. Regulation helps the big and hurts the small. Compliance infrastructure is a sales advantage for a firm with 15,000 people and a burden that can put a bedroom startup in legal jeopardy; meanwhile European governments quietly race to adopt AI and will likely be forced to soften their own rules. The football-team principle. If you're allowed eleven humans on the pitch and unlimited robots, you field eleven humans plus robots. AI supersizes people rather than replacing them – and at TheBigWord, that's a stated commitment. Chapters: 00:04 The Evolution of Machine Translation 02:58 Moses, Hybrid Models and the Road to Neural MT 07:46 How Google Changed the Accuracy Game 13:17 The Role of Human Linguists in AI 17:30 Building the Orchestration Layer 20:21 The Real Cost of AI Tokens and Interpreting 22:36 The Economic Value of Immigration and Language Access 26:43 Cultural Nuances in Translation 31:05 Navigating Global AI Regulation 39:42 Regulatory Challenges in AI and Translation Hyperlinks: TheBigWord Josh Gold on LinkedIn Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3) First Public Reading, Romy, Roby and the Secrets of Sleep (2/3) First Public Reading, Romy, Roby and the Secrets of Sleep (3/3) AI Snacks with Romy and Roby @romyandroby “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book Substack

Remember Napster — the file-sharing site the entire music industry tried to kill? It's back, and it isn't a streaming service anymore. In this episode, Anastassia talks with Sam Huber, the physicist who once ran Lewis Hamilton's engine strategy at Mercedes Formula One, built an in-game advertising company from zero to 120 employees, rode the metaverse wave with Landvault, and now leads Napster's enterprise business from Dubai. Today's Napster builds «streaming intelligence»: teams of specialized, avatar-fronted AI agents — more than 20,000 of them — that companies hire like employees, onboard with their own data, and supervise like a team. Sam explains why the interface, not the intelligence, is where AI will be won; why your company's first-party data holds 90% of the value; and why the future belongs to «elastic organizations» that scale their workforce up and down like cloud computing. The conversation closes with a message every student and every worried employee should hear: AI should augment you, not replace you — and the people at risk are not those whose jobs meet AI, but those who never learned to use it. Key Takeaways: The interface is the frontier, not the model. Foundation models will keep improving without Napster's help — the huge, underexploited delta is how humans engage with AI. Hire AI agents like employees, not software. Napster's 20,000+ agents are trained on domain-specific curricula, then onboarded with your company's documents, APIs, and live data feeds — like a new hire on their first day. Ninety percent of the value is your own data. A job is not a task. Jobs begin with judgment (deciding what to do) and end with judgment (checking it was done right); AI does the middle. Abundance beats headcount-cutting. Doing $1m with 10 people instead of 20 is a capped, short-sighted strategy. The interesting question: how do you do $5–10m with more agents AND more humans to manage them? The elastic organization. What cloud computing did for servers, AI agents do for the workforce. Fragmented «Frankenstein» agents fail. A marketing agent that doesn't know what the finance agent is doing optimizes one silo and may hurt the business. Agents need organizational context — teams, reporting lines, who sits where — to be labor, not software. The Gulf has deadlines; Europe has debates. Government AI strategies in the region set automation targets by year, creating urgency at chairman and sovereign-fund level to redesign operating models — not to buy point solutions. Meanwhile, Europe's demographic crunch makes automation a necessity, not a choice. Chapters: 02:20 — Sam Huber's Journey: From Physics to AI 08:32 — The Evolution of Napster and AI Integration 14:02 — Humanizing AI: The Role of Specialized Agents 19:07 — Data Sourcing and Privacy in AI Solutions 21:10 — The Value of First-Party Data 22:54 — Governance and Transparency in AI 24:10 — AI as an Augmenter, Not a Replacer 28:46 — Elastic Organizations and Creative Industries 33:18 — The Future of HR in the Age of AI 39:42 — Education and AI: Preparing for the Future Hyperlinks: NapsterSam Huber on LinkedIn Sam Huber on Instagram Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3) First Public Reading, Romy, Roby and the Secrets of Sleep (2/3) First Public Reading, Romy, Roby and the Secrets of Sleep (3/3) AI Snacks with Romy and Roby @romyandroby “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book Substack

Summary: Anastassia is joined by Professor Rae Muhlstock for a close reading of Kazuo Ishiguro’s Klara and the Sun — a Nobel laureate’s quiet, devastating novel narrated by an AI, and soon a major film directed by Taika Waititi, starring Jenna Ortega and Amy Adams. Dr. Rae Muhlstock is a Teaching Professor in the Writing and Critical Inquiry Program at the University at Albany (SUNY). She holds a PhD from the University at Buffalo, specializing in 20th and 21st century fiction, film, and narrative theory, and teaches a beloved course on AI, science fiction, and philosophy. Her published work spans the Classical myth of the labyrinth. She chairs the annual WCI Film Festival in Albany. This episode is part of our ongoing series “AI Through the Lens of Storytellers,” in which Anastassia and Rae discuss portrayals of AI in books and movies — from Ex Machina and Her to Data in Star Trek and After Yang — because these stories shape our expectations about AI technologies. And because art is an experiment: a way for thinkers to explore future scenarios that can’t be tested in labs. Key Takeaways: Klara is three archetypes in one. The perfect, willingly self-erasing servant; the worshipper who builds a private theology around the Sun; and the doppelgänger — a double whose purpose raises an ethical horror the novel only slowly reveals. Sacrifice may be a benchmark of life. Klara’s sacrifice is arguably the purest in the book because it comes from love without societal pressure. The most human character is the AI. Ishiguro’s humans are trapped in systems and status; it is Klara who models feeling, connection, and grace. AI in the real world is drifting toward religion. With institutions of enormous power, apostles, heretics, prophecies of doom, and millions of followers, today’s AI discourse mimics the structure of a religious movement. Keep your agency. Today’s AI mimics; it does not understand, reason, or learn from counterfactuals. Fiction is how thinkers experiment. Scientists experiment in labs; philosophers and artists experiment through stories. Chapters: 00:04 — Welcome to the storytellers series: From Ex Machina, Her, Star Trek, and After Yang to Ishiguro’s quietest, most devastating AI. 02:24 — Klara goes to Hollywood: The upcoming Taika Waititi film with Jenna Ortega and Amy Adams — and what a Nobel laureate’s AI meditation becoming a studio production says about our cultural moment. 03:19 — The plot (and why it’s the least interesting thing): Rae on Ishiguro’s “quiet science fiction,” first-person narrators, and the reader who always knows both more and less than Klara. 07:34 — Three faces of Klara: The perfect slave, the worshipper of the Sun, and the doppelgänger — archetypes, ethical horror, and a hierarchy of machines. 12:44 — Sacrifice as a sign of life: What Klara, Data, and The Creator suggest about love, agency, and the line between performing and being. 19:14 — Is AI becoming a religion?: Churches, apostles, heretics, and cults in today’s AI discourse — and Capaldi, the man who doesn’t believe in the human soul. 29:27 — Keep your agency: Why AI mimics rather than understands, the missing math of counterfactuals and causality, and the “chosen ones” narrative of AI builders. 36:42 — Fiction as an experiment — and the classroom: Checking ChatGPT’s math, cheating smart vs. cheating dumb, hidden environmental costs, and why we must always ask why. Hyperlinks: Rae Muhlstock University at Albany faculty profile LinkedIn Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3) First Public Reading, Romy, Roby and the Secrets of Sleep (2/3) First Public Reading, Romy, Roby and the Secrets of Sleep (3/3) AI Snacks with Romy and Roby @romyandroby “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book Substack

Dr. Marschall Runge returns for a second conversation, this time about the people who will practice medicine next.  Marschall S. Runge, MD, PhD, is a practicing cardiologist who spent a decade as executive vice president for medical affairs at the University of Michigan, dean of the Medical School, and CEO of Michigan Medicine — running one of America's largest academic health systems while continuing to see patients. He earned his PhD in cardiovascular molecular biology at Vanderbilt and his MD at Johns Hopkins, where he also completed his internal medicine residency, followed by a cardiology fellowship at Massachusetts General Hospital. He now leads a Michigan cohort combining epigenetic aging clocks, mitochondrial inheritance, cardiorespiratory fitness, and chronic inflammation in a single dataset. He has authored more than 195 peer-reviewed papers and holds five healthcare patents. A third nonfiction book, on epigenetic clocks and biological aging, is in development with Forbes Books, alongside a forthcoming Substack (The Longevity Switch) and a University of Michigan MOOC on epigenetics and aging. Chapters 00:00 Introduction and guest overview 01:12 AI's role in healthcare transformation 04:01 AI as a valuable assistant in medicine 05:08 Redefining medical curriculum for AI era 09:34 Openness of medical institutions to AI 11:47 Changing accreditation and certification with AI 16:08 Leadership in AI adoption in healthcare 18:39 International perspectives on AI in medicine 19:13 AI's impact on radiology and diagnostics 21:46 AI in clinical workflows and operations 24:49 The evolving role of nurses with AI 30:37 Digital twins and clinical trials 33:18 Transition period and future outlook Hyperlinks: The Great Healthcare Disruption: Big Tech, Bold Policy, and the Future of American Medicine (Forbes Books, 2025) — USA TODAY Best-Seller and Global Book Awards Gold Medal winner. An insider's account of the forces reshaping American healthcare.  Coded to Kill (Post Hill Press, 2023) — a techno-medical thriller. Amazon author page Dr. Runge Website OpenEvidence — the physician-only clinical AI Runge calls superb  University of Michigan Medical School U-M Center for Academic Innovation  Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (2/3)  First Public Reading, Romy, Roby and the Secrets of Sleep (3/3)  AI Snacks with Romy and Roby @romyandroby  “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book Substack

Summary AI is extraordinary at finding answers. But who finds the problem? In the 4th of our Beyond Human series, Anastassia and Matthias Röder go after a question: can an AI formulate a genuinely new problem, or is it forever confined to solving the ones we hand it? Dr. Matthias Röder is a music and technology strategist based in Salzburg, and one of the most consistently original thinkers on creativity and machines that we know. He is a former Managing Director of the Eliette and Herbert von Karajan Institute and a board member of the Karajan Foundation, as well as a trustee of the Mozarteum Foundation. He is co-founder and managing partner of The Mindshift, a consultancy on creative leadership and innovation strategy. Matthias directed the Beethoven X project, the AI-assisted completion of Beethoven's Tenth Symphony, work that earned an Effie Bronze. He founded the Karajan Music Tech Conference in 2017 and launched the Classical Music Hack Day series in 2013. He holds a PhD in music from Harvard University and is an alumnus of the Mozarteum University Salzburg. His work has been recognized with the "Game Changer" award of the Salzburg Chamber of Commerce. Chapters 00:01 Welcome, and the question behind the episode: can AI formulate problems, not only answer them? 01:48 In music, the problem is the composition — why framing the question is the hardest creative act 03:17 35 million papers, 4,000 a day: PubMed, AlphaFold, and what no human can read 05:58 GNoME and 2.2 million crystals — when the hypothesis machine outruns the laboratory 09:09 World models, physical intuition, and the questions AI does not know to ask 11:36 A map of all unsolved problems: knowledge graphs as an orchestration engine 15:16 Progress without consensus — why agreement may be the most expensive thing we do 17:24 Theory of mind, Belle and Rock Experiment, and why your chatbot keeps telling you that you are a genius 24:20 Context contamination, hallucinations, and professional AI hygiene 30:50 Organizational elasticity, the 2×2 of known and unknown, and three papers worth your time Papers mentioned in the episode: Towards Scientific Discovery with Generative AI: Progress, Opportunities and Challenges — a survey of AI across the full scientific discovery cycle, from literature analysis and brainstorming through theory refinement, experimental design and data-driven discovery. The Future of Fundamental Science Led by Generative Closed-Loop Artificial Intelligence — an ambitious argument that AI is approaching the ability to close the scientific loop autonomously, from hypothesis generation through experimental design to validation. Machine Learning as a Tool for Hypothesis Generation — a procedure for generating hypotheses from high-dimensional behavioural data and testing them on held-out human subjects, grounded in economics and behavioural science. Also referenced: DeepMind's AlphaFold and GNoME; AI Feynman and the rediscovery of fundamental physics equations from data alone; Yann LeCun on world models; Jorge Luis Borges, The Library of Babel. Hyperlinks: Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3) First Public Reading, Romy, Roby and the Secrets of Sleep (2/3) First Public Reading, Romy, Roby and the Secrets of Sleep (3/3) AI Snacks with Romy and Roby @romyandroby “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book Substack

In this episode of AI Snacks with Romy & Roby, Anastassia Lauterbach welcomes Paul Loeb — entrepreneur, musician, audio engineer, and the founder of DropTrack, the world's first AI-powered music promotion platform. The conversation moves fluidly between Paul's personal journey from bedroom DJ and record label founder to tech executive at Beats by Dre, Apple, GitHub, and Ticketmaster, and the real-world mechanics of how DropTrack is changing what it means to be an independent artist in the age of AI. At its heart, this episode is about democratisation: of music promotion, of AI literacy, and of access to audiences that were previously reachable only by those with money, connections, or industry clout. Paul and Anastassia dig into the tension between AI as a creative threat and AI as an empowerment tool, explore the copyright battles reshaping the music industry, and discuss how independent artists and major labels alike need to rethink their strategies — before the window of opportunity closes. The episode closes with a call to action for conservatories and music schools: AI literacy is no longer optional for the next generation of artists. Chapters 00:00 Introduction to AI Snacks and the episode topic 00:55 The journey of music from creation to listener 01:54 Paul Loeb's background and path into music and tech 03:37 Challenges of music promotion and the birth of DropTrack 07:16 How DropTrack uses AI to match music with opportunities 11:01 Using AI for music analysis and readiness assessment 12:43 Matching music to labels, playlists, and licensing opportunities 15:21 The role of human curation and AI in discovering new music 18:30 Global reach and different strategies for artists 21:49 Disruption of traditional music industry giants 24:38 Copyright, metadata, and royalties in AI-driven music 28:24 Training AI models with existing music and copyright concerns 33:31 Common mistakes and misconceptions about AI in music 36:29 AI literacy as a crucial skill for young artists 39:27 Closing remarks and the future of AI in music About the Guest Paul Loeb is the Founder and CEO of DropTrack, the world's first AI-powered music promotion platform. He has over 20 years of experience spanning music and tech, with roles at Beats by Dre (acquired by Apple), GitHub (Microsoft), and Ticketmaster/Live Nation. He founded No Ego Records in 2010, signing over 50 artists and releasing more than 100 tracks. A USC graduate and Certified Pro Tools Expert trained under Steely Dan producer Roger Nichols, Paul also produces and DJs electronic music under the alias Really Cute Cats. He is based in Los Angeles, California.  About the Podcast AI Snacks with Romy & Roby is the AI literacy podcast hosted by Dr. Anastassia Lauterbach, Founder and CEO of AI Edutainment GmbH. Each episode transforms complex ideas in artificial intelligence into something accessible, tangible, and actionable — for educators, executives, artists, and curious humans everywhere. The podcast is available on Spotify, Apple Podcasts, and all major platforms.

Summary Anastassia sits down with Tony Falco of Hydrolix to explore one of the least-discussed but most consequential challenges in the AI era: CDN and log data. Every click, stream, and page interaction generates hundreds of log files per second. For nearly three decades, that data has been thrown away — not because it lacked value, but because storing it was prohibitively expensive. Tony brings over 30 years of experience building the internet's infrastructure, from his early days at Akamai helping Fortune 500 companies scale their websites for Super Bowl traffic, to Silicon Valley Data Science (acquired by Apple), to co-founding Hydrolix — a platform purpose-built to make petabyte-scale log data affordable, queryable, and AI-ready. The conversation moves from the very origins of the commercial internet, through the evolution of distributed databases and CDN technology, to the urgent present: a moment when AI is simultaneously generating new value from log data and being weaponised against the websites that produce it. Tony and Anastassia explore the governance gap around bots and agents, the coming convergence of cybersecurity and data strategy, and why the biggest opportunities in AI may still be hiding in data that nobody has thought to explore. Key Takeaways: Log files are the dark matter of the internet. Every interaction with a website generates hundreds of log files per second. For 26+ years, this data has largely been deleted after 90 days because of cost — but it contains the entire behavioural story of a website's audience. The economics breakthrough unlocks the AI opportunity. Hydralix was built to reduce CDN log storage costs by up to 90%. But when given the choice of cutting costs or keeping data longer, 99% of customers choose to keep the data — because the insights it holds are more valuable than the savings. AI is hungry for exactly the data that has been thrown away.  AI is also the new threat on the attack surface. Bots and agents are increasingly using AI to mimic human behavior — bypassing detection systems that relied on the predictable signatures of scripted traffic.  Intent is the new frontier. Hydralix focuses on inferring the intent of website visitors — human or bot. Normal intent (browsing, comparing, buying) can be profiled; variance from it flags bad actors. Sophisticated attackers are patient. If they know your data retention window is 90 days, they will make their next move on day 91.  The bot/agent governance gap is real and growing. Businesses have DUNS numbers. Websites have no equivalent system to identify or authenticate agents.  Don't accept the current narratives about AI. Tony's core message: it is very early. The biggest opportunities have not been taken. The biggest decisions have not been made. Curiosity and drive matter more than credentials right now. Chapters: 00:01 Introduction, Tony’s Journey 05:30 From CDNs to NoSQL to Gen AI: The Through-Line 08:47 Data explosion and the Log File problem 14:21 The Marianas Trench of Data 15:57 Understanding niche markets 16:35 Competitive landscape of data providers 19:43 Real-time correlation as the competitive advantage 21:07 Intent Analysis and Bot Detection 23:56 The Governance Gap: Bots, Agents and Identity 29:26 Advice for Young People entering the field 32:06 Machine-to-machine, bot-to-bot workflows and communication 32:50 Prometheus Moment of AI: Building tools without fully understanding the consequences 47:14 Closing Hyperlinks: Tony Falco Hydrolix Website Anastassia Lauterbach - LinkedIn First Public Reading, Romy, Roby and the Secrets of Sleep (1/3) First Public Reading, Romy, Roby and the Secrets of Sleep (2/3) First Public Reading, Romy, Roby and the Secrets of Sleep (3/3) AI Snacks with Romy and Roby @romyandroby “Leading Through Disruption” AI Edutainment The AI Imperative Book Romy & Roby Book Substack