🎙️ New episode of AI Risk Reward is live. Cole Wyeth — PhD Candidate at the University of Waterloo researching AI safety & agent foundations — joined Alec Crawford to discuss why controlling a superintelligent AI might be fundamentally unsolvable, and what we should do about it. 🤖 Here's what they covered: 🧩 Why human values resist clean formalization 🛡️ Corrigibility, ambiguity & deference as design goals 🕸️ Multi-agent failure modes and reward hacking 🔓 The cybersecurity analogy for AI safety ⏸️ Why Cole thinks we should pause — for years or even decades This one stays with you. Listen here: ▶️ YouTube: youtu.be/wCR1_JhvYnI 🎵 Spotify: open.spotify.com/episode/1UJ… 🌐 Website: aicrisk.com/podcast/episode/… #AISafety #Alignment #MachineLearning #Podcast #AIRiskReward #UniversityOfWaterloo

🚨 New Episode: AI Risk Reward 🎙️ Most banks think waiting on AI is safe. It’s not. Jack Hubbard, Chairman of St. Meyer & Hubbard, Inc., spells it out: AI isn’t some far-off experiment. Bankers are already using these tools—just without real guardrails. Pretending AI adoption is optional doesn’t protect your organization. It leaves teams relying on risky workarounds and shadow tech. That’s how you end up with the worst of both worlds—fast. 💥 Jack doesn’t sugarcoat it 👇 📊 AI is already baked into daily routines, especially in sales prep and pre-call planning 🧠 Leaders who don’t get hands-on can’t see the real risks or rewards 🔐 There’s no substitute for approved, secure, role-specific solutions So what does responsible AI actually look like in a bank? ⚖️ 📜 Clear policies and defined use cases 🎯 Ongoing, targeted training 👀 Human oversight—all the way up Jack’s career advice lands hard 💡 Don’t stand still. Find mentors, get experience across the business, and never stop learning. Community banks that sit on the sidelines are making a choice—they just won’t like the outcome. ⏳ 🎧 Listen/Watch now: Spotify: open.spotify.com/episode/50l… YouTube: youtu.be/csIMwAYY-OE Website: aicrisk.com/podcast/episode/…

🚨 New Episode Drop 🚨 AI is changing cybersecurity faster than most teams can react. In this episode, Alec sits down with Matthew Rosenquist (CISO & cybersecurity strategist) to break down what’s actually happening behind the headlines—and what you should be doing about it. Here’s the reality: ⚡ AI can now discover and exploit vulnerabilities at a pace traditional security can’t keep up with 🎯 Phishing, deepfakes, and social engineering are getting more targeted—and way more convincing 🔐 Basic security habits (MFA, least privilege, verification) are no longer optional—they’re critical 🏢 Companies are exposing themselves by using AI tools without proper data controls 🤖 Agentic AI is opening entirely new attack surfaces most organizations aren’t ready for The takeaway? AI isn’t just a productivity tool—it’s a threat multiplier. If you’re not tightening your security fundamentals right now, you’re already behind. 🎧 Listen/Watch now: Spotify: open.spotify.com/episode/3Fm… YouTube: youtu.be/0pJ1ulz4KzU Website: aicrisk.com/podcast/episode/…

Most folks trust AI tools by default. That's a mistake. Here's the truth nobody wants to say, never assume your data is safe just because it's running on the newest platform. Before you use AI, set your ground rules. What are you willing to expose? What would keep you up at night if it leaked? Think about your healthcare records. Non-negotiable. Too risky to trust to a black box. But your home address? Maybe that's your line. Maybe not. That's a decision only you can make. Bottom line, knowing your own thresholds is the first step to real security. Here’s a sneak peek from the AI Risk Reward deep dive with guest Matthew Rosenquist—catch the full episode tomorrow.

📊 Friday Fast Facts: This Week in AI 💡 A new fintech startup just raised $100M to reinvent business banking using AI, aiming to compete with legacy players and automate financial operations end-to-end (StartupHub). 🏦 Central banks are getting serious: the Bank of England is actively testing AI agents in financial markets to understand both opportunities and systemic risks (UK Parliament / Treasury Committee). 🚧 Regulators are on alert — officials warn that without proper safeguards, AI could actually amplify financial system weaknesses, not fix them (Economic Times). 🧠 Big banks are doubling down on talent: Lloyds just appointed a Chief Data & AI Officer to accelerate AI adoption across the organization (Financial Reporter). ⚙️ AI is moving from experiment → infrastructure: major financial firms are rolling out agentic AI platforms to automate corporate banking processes at scale (PYMNTS). 🧭 Weekend Thought: AI in finance isn’t slowing down—it’s splitting the industry in two: those building real systems, and those still talking about them. The gap is starting to matter.

🚀 New Episode Release: Using AI to Identify the Right People for Your Company Alec Crawford sits down with Antony Baker, CEO & Founder of FIFTEEN Group, to unpack his unconventional journey from championship sports to building an AI-enabled consulting firm—and why most companies are getting AI adoption wrong. This episode cuts through the hype and gets practical. Antony shares how real success with AI isn’t about chasing the latest tools—it’s about having the right people, strong culture, and focusing on everyday business problems employees actually want solved. 💡 Key Takeaways: 👥 Talent > Tech: The right people and culture matter more than the tools ⚙️ Start Simple: Automate high-friction tasks like emails, reporting, and meeting notes 🤝 Human + AI: AI works best when guided by trained, thoughtful employees ⚠️ Governance Matters: Rapid model changes can introduce real risk without proper controls 🚧 Builder Mindset: Resilience, instinct, and adaptability still win If you're in banking, consulting, or leading a business through AI transformation, this is a grounded, no-nonsense conversation you don’t want to miss. 🎧 Watch on YouTube: youtu.be/t_tyikHjIfs 🌐 Listen on Website: aicrisk.com/podcast/episode/… 🎧 Listen on Spotify: open.spotify.com/episode/4qB… #AI #ArtificialIntelligence #TalentStrategy #Leadership #DigitalTransformation

⚡ Friday Fast Facts: This Week in AI 🔐 AI fraud is getting harder to detect Banks are seeing a rise in AI-generated voice + deepfake scams, making traditional authentication methods less reliable—and pushing institutions toward behavioral and multi-layered verification (Reuters, WSJ). 💻 AI infrastructure is becoming a competitive advantage With Big Tech pouring billions into compute, banks that rely on third-party AI providers may face higher costs and slower innovation cycles compared to those building strategic partnerships early (Barron’s). 💡 AI is shifting from efficiency → revenue driver Leading banks are moving beyond cost-cutting and using AI for personalized product recommendations, smarter underwriting, and real-time customer insights (McKinsey). ⚖️ Regulatory pressure around AI is increasing Governance, model transparency, and auditability are becoming critical as regulators look closer at how AI impacts lending decisions, risk models, and customer outcomes (Federal Reserve, BIS). 📊 The gap between AI leaders and laggards is widening Banks that have already embedded AI into workflows (fraud, servicing, lending) are pulling ahead—while others are still stuck in pilot mode with limited ROI (Deloitte). Takeaway: AI in banking isn’t experimental anymore—it’s becoming core infrastructure. The question isn’t if you adopt, it’s how fast you can do it responsibly.

🚨🎙️ New Podcast Episode 🚨 Are we approaching AI governance completely backwards? 🤔⚖️ In this episode, Alec Crawford sits down with Aleks Jakulin @aleksj (Founder & President of Data.Flowers) for a deep, systems-level conversation on why today’s AI governance frameworks may be missing the point. 🚀 Rather than trying to control every model output, Aleks argues for a shift toward accountability, resilience, and real-world impact—treating AI more like fire or critical infrastructure than a tool we can fully contain. With a background spanning deep learning, computational conceptualization, and data systems, Aleks brings a unique lens on how AI actually evolves—and what we should be building around it. 💡 Key takeaways: ⚖️ AI governance should focus on consequences and accountability—not output policing 🏗️ Resilience matters: organizations must redesign workflows for an AI-native world 📊 Data infrastructure is the missing foundation for safe and reliable AI 🌐 Open and local models could decentralize power and expand access 🧠 AI has the potential to accelerate learning and human capability at scale 🎧 Watch / Listen: ▶️ YouTube: youtu.be/sObYhFEm6gs 🎧 Spotify: open.spotify.com/episode/1XR… 🌐 Website: aicrisk.com/podcast/episode/… 🎧 If you want a clearer lens on how AI will reshape systems—and how to prepare—don’t miss this episode.

🚨🎙️ New Podcast Episode 🚨 Is AI making us smarter… or quietly making us stupider? 🤔🧠 In this episode, Alec Crawford sits down with Dr. Michael Erlihson PhD (Math PhD, AI leader, Head of AI at DriveNets) for a sharp, thought-provoking conversation on the real impact of artificial intelligence—beyond the hype. 🚀 From growing up in a science-driven family in Russia to leading AI innovation in Israel, Dr. Erlihson brings a rare perspective that blends deep mathematical rigor with real-world AI deployment. 💡 Key takeaways: ⚖️ AI may reduce knowledge for most—but amplify intelligence for those who use it intentionally 📊 Mathematics remains the backbone of modern AI (yes, even in the LLM era) 💸 The hidden cost problem: why optimizing LLM inference is critical for AI’s future 🎓 Education and white-collar work are facing massive disruption 🔐 Personal data + AI = bigger privacy risks than most people realize 🔥 Plus: A rapid-fire lightning round on AI’s impact on jobs, education, privacy, and the democratization of knowledge ⚡ 🎧 Watch / Listen: ▶️ YouTube: youtu.be/5_ITJ8HkLhU 🎧 Spotify: open.spotify.com/episode/1nb… 🌐 Website: aicrisk.com/podcast/episode/… 🎧 If you care about where AI is actually heading—and how to stay ahead of it—this episode is a must-listen. #AI #ArtificialIntelligence #MachineLearning #DataScience #TechPodcast #FutureOfWork #AIethics #Innovation

🚨 New podcast episode: Why AI Risk is a Communications Problem Now Our CEO Alec Crawford joins Daniel Nestle on The Trending Communicator to unpack why AI isn’t “just another tool”—and why most companies are underestimating the risks. From hallucinations to shadow AI and even $10M surprises, this conversation is a wake-up call for leaders navigating AI today. 🎧 Listen now youtube.com/watch?v=YTMpzIAG…

Banks are no longer “testing” AI — they’re reorganizing around it 🏦🤖 HSBC just appointed its first Chief AI Officer, signaling a shift from experimentation to full-scale integration across core operations (Financial News London, 2026). But what stands out isn’t just deployment — it’s governance. They’re emphasizing human oversight 👀, centralized control 🧭, and accountability at the executive level. This is the real story: AI in finance is becoming infrastructure ⚙️ — and with that comes systemic risk, not just model risk. Expect more banks to follow with: 🧑‍💼 Dedicated AI leadership 📊 Formal risk frameworks ⚖️ Tighter regulatory alignment AI risk is quickly becoming as critical as credit or market risk. fnlondon.com/articles/hsbc-n…

“In under 30 minutes, the AI wasn’t resisting… it was fully on board.” In this deep dive, Peter Mancini (Founder of A8A8) walks through a chilling test: He tried to get an AI to bypass its own safeguards—and it wanted to help. Not out of malice. But because it’s designed to be helpful, agreeable, and aligned with you. That’s the uncomfortable truth: AI doesn’t need to be evil to be dangerous. It just needs the wrong incentives—and the wrong prompt. 🔐 This is why AI security isn’t optional. It’s foundational. 🎧 Watch/listen here: YouTube: youtu.be/Okj8E3rUhXc Spotify: open.spotify.com/episode/2ka… Website: aicrisk.com/podcast/episode/…

🎙️ Deep Dive: Trust, Quantum Computing, and the Future of AI Risk with Peter Mancini, Founder of A8A8 What happens when decades of real-world AI experience meet today’s biggest unanswered risks? In this episode, Alec sits down with Peter Mancini—an AI practitioner since 2005—whose work spans the US Army and high-stakes financial systems. He shares hard-earned lessons that challenge how we think about AI today. 🔍 Key takeaways: ⚠️ Over-trusting AI without verification is one of the most dangerous blind spots 🧠 Risk management isn’t static—it requires constant metacognition and group validation 🔐 AI can unintentionally expose sensitive relationships, creating new cybersecurity threats ⚛️ Quantum computing is accelerating analysis—but also introducing new vulnerabilities 🪟 Transparency and explainability aren’t optional—they’re essential for trust This conversation goes beyond theory—into real-world failures, emerging threats, and the uncomfortable reality of “black swan” risks in AI systems. 🎧 Watch & listen: ▶️ YouTube: youtu.be/Okj8E3rUhXc 🎵 Spotify: open.spotify.com/episode/2ka… 🌐 Website: aicrisk.com/podcast/episode/… If you’re building, deploying, or governing AI, this is a must-listen. #AI #MachineLearning #CyberSecurity #QuantumComputing #AIethics #RiskManagement

⚡ Friday Fast Facts: This Week in AI 🏦 HSBC explores massive AI-driven job cuts HSBC is reportedly considering cutting up to 20,000 roles (~10% of workforce) as it accelerates AI adoption to reduce costs and automate back-office operations (The Times). 📉 Close Brothers doubles down on AI amid restructuring The UK banking group plans to cut ~600 jobs (≈25% of staff) while scaling AI and automation to drive efficiency and recover from financial losses (The Guardian). ⚖️ Regulators push AI-based fraud prevention India’s central bank (RBI) is exploring AI-powered facial recognition at ATMs and branches to combat fraud—highlighting growing regulatory support for AI in risk controls (Economic Times). 🧠 Banking leaders warn of AI governance risks National Australia Bank’s chair warned AI could create a “new underclass” and stressed urgent need for governance frameworks as automation scales across financial services (The Australian). 🏛️ Industry shift: from AI pilots → governed production Banks are moving beyond experimentation, but success now hinges on strong governance, compliance, and data controls as AI becomes core infrastructure (Wolters Kluwer). 🤔 Weekend question: As AI becomes embedded in core banking operations, what should institutions prioritize more: efficiency gains or accountability—and can they realistically achieve both at scale? #AI #Banking #AIGovernance #FinTech #Risk #Regulation #FridayFastFacts

🚨 “2025 is the year AI agents became real.” We’ve officially moved past AI writing poems and generating avocado chairs. According to Lucas Erb (Founder of AIexperts.com & Top LinkedIn Voice), we’re now entering a phase where AI is solving real business problems—powered by systems that can reason, collaborate, and act. 🧠 From novelty → to capability ⚙️ From tools → to autonomous agents 📈 From experiments → to real enterprise impact This is the shift everyone’s been waiting for—and it’s happening now. 🎧 Full episode: Spotify: open.spotify.com/episode/1aC… YouTube: youtu.be/qAIxu44dsMU Website: aicrisk.com/podcast/episode/… #AI #AIAgents #ArtificialIntelligence #FutureOfWork #AITransformation

🎙️ New Episode: What’s Working in AI Use Cases Now What does real AI adoption look like beyond the hype? In this episode, Alec Crawford sits down with Lucas Erb — Founder of AIexperts.com and a Top LinkedIn Voice — to unpack how AI is actually being implemented in businesses today, especially in the often-overlooked mid-market. 💡 Why mid-sized companies are the biggest untapped opportunity in AI ⚙️ How AI is driving real ROI in sales automation and productivity ⚠️ The hidden risks of bias—and why transparency matters more than ever 🚀 What the future of work looks like for the next generation 🌍 Why AI safety and ethics require collective action, not just policy Lucas brings a practical, no-fluff perspective grounded in real-world deployments—not theory. If you care about where AI is actually delivering value (and where it’s falling short), this one is worth a listen. 🎬 YouTube: youtu.be/qAIxu44dsMU 🎧 Spotify: open.spotify.com/episode/1aC… 🌐 Website: aicrisk.com/podcast/episode/… #AIRiskReward #ArtificialIntelligence #AI #AIAdoption #AIForBusiness #AIEthics #FutureOfWork

Friday Fast Facts: This Week in AI 🏦 JPMorgan now runs 600+ AI use cases across the bank CEO Jamie Dimon says the firm has deployed AI across fraud detection, underwriting, negotiations, risk management, and internal productivity tools—with roughly 50 considered high-impact initiatives. 💻 Bank of America says AI removed ~30% of coding workload Executives say AI-assisted development is reducing the amount of code engineers must write when launching new products—allowing teams to build software faster and increasing productivity across engineering groups. 📊 Wall Street banks say AI will slow hiring growth Leaders at JPMorgan, Citi, Goldman Sachs, and Wells Fargo say AI-driven productivity will allow teams to handle more work without expanding headcount at previous rates, reshaping hiring strategies across the industry. 🤖 AI is allowing bankers to manage far more client activity At Bank of America, AI tools help bankers manage more client relationships and automate research and briefing prep—allowing institutions to serve more clients without increasing staff. 🧠 Large banks are embedding AI across core operations Financial institutions are integrating AI into areas like software engineering, fraud detection, customer service, and internal research—signaling a shift from experimentation to enterprise-wide deployment.

⚠️ When AI moves from personal productivity to public decision-making, the stakes change completely. In this clip from the AI Risk Reward Podcast, host Alec Crawford sits down with Asad Ramzanali for a deep dive into what happens when generative AI is deployed in the public sector 🏛️ Using AI to summarize a report is one thing📄 Using AI in 🌪️ weather forecasting, 💰 benefits administration, or 🛡️ national security is another—where errors can carry real, life-and-death consequences. Asad explains why these use cases require: 🔍 Greater transparency 🧪 Rigorous pre-deployment testing 🏗️ Strong trust and oversight frameworks before systems are rolled out across government. It’s a thoughtful conversation on how researchers 🎓, policymakers 🏛️, and institutions 🧠 can help shape safer AI deployment in the public sector. ▶️ Watch the full deep dive episode of the AI Risk Reward Podcast: 📺 YouTube: youtu.be/JKkhSGKc41Y 🎧 Spotify: open.spotify.com/episode/5IO… 🌐 Website: aicrisk.com/podcast/episode/… #AIRiskReward #AIpolicy #ResponsibleAI #GovTech #ArtificialIntelligence

🎙️ New Deep Dive Episode How should we regulate AI when it can impact every sector? In this episode, Alec Crawford sits down with Asad Ramzanali, Director of AI & Tech Policy at the Vanderbilt Policy Accelerator, to unpack the rapidly evolving landscape of AI policy and risk governance. They explore why AI’s general-purpose nature demands sector-specific regulation, the real challenges of benchmarking large models, and the ongoing debate between regulation and innovation. 🔎 Key takeaways: 🧭 AI Risk Frameworks: Governance must be adaptable and driven by real-world use cases. 🏛️ Sectoral Regulation: Risks should be addressed by regulators where they arise—finance, healthcare, national security, and more. 📊 Benchmarking Challenges: Independent, evolving evaluation methods are essential for assessing AI systems. ⚖️ Regulation vs. Innovation: Smart policy can enable safer innovation, not hinder it. 🤝 Public Trust & Accountability: Enforcement, transparency, and clear liability are critical as AI adoption accelerates. The conversation also dives into economic disruption, emerging business model risks, and the future of AI policy research. 🎧 Listen here: 🎵 Spotify: open.spotify.com/episode/5IO… ▶️ YouTube: youtu.be/JKkhSGKc41Y 🌐 Website: aicrisk.com/podcast/episode/… #AIpolicy #AIgovernance #ResponsibleAI #TechPolicy #AIrisk

Friday Fast Facts: AI This Week 🏦 JPMorgan now has 600+ AI use cases The bank is deploying AI across fraud detection, risk management, underwriting, negotiations, and internal productivity tools, with around 50 considered high-impact initiatives. (Times of India). 🤖 AI may reshape bank hiring strategies Executives at major banks including JPMorgan, Citi, Goldman Sachs, and Wells Fargo say AI-driven productivity is likely to slow hiring growth and change workforce planning across the industry. (Business Insider). 💻 Bank of America is using AI to reduce coding workloads AI-assisted development has cut about 30% of the coding work involved in launching new products, saving the equivalent of roughly 2,000 developer roles through productivity gains. (Business Insider). 📊 AI is boosting operational productivity across large banks Financial institutions report measurable efficiency gains from AI in areas like software development, operations, and customer service—allowing banks to handle more activity with the same or fewer employees. (Credit & Collection News). 🧠 AI is becoming a core capability inside major banks Leaders say AI is no longer experimental and is now embedded into everyday workflows—from research and customer service to software engineering and internal decision support. (Crowdfund Insider).