538. Methodology Over Model

In this episode, Transform NOW co-hosts Brad Hairston and Michael Marchuk banter about some interesting recent news headlines in the world of automation and AI. News stories covered in this episode:• What a $1.2 Billion Lesson Means for Your AI Strategy: Automation Without Accountability https://www.jdsupra.com/legalnews/what-a-1-2-billion-lesson-means-for-6095787/ • Companies Want More Entry-Level Workers. They Don’t Want Beginners https://www.inc.com/bruce-crumley/companies-want-more-entry-level-workers-they-dont-want-beginners/91429033 • Musk Says Grok Bot’s Costs Were ‘Unsustainably High’ Weeks After Touting Rapid Growth - https://finance.yahoo.com/technology/ai/articles/musk-says-grok-bot-costs-111916668.html • Anthropic warns that it’s very close to RSI capability, aka self-building AI - https://techcrunch.com/2026/11/02/anthropic-warns-that-its-very-close-to-rsi-capability-aka-self-building-ai/ • A Look at How AI Is Working Out in the Workplace https://www.gsb.stanford.edu/insights/look-how-ai-working-out-workplace • Google faces mounting backlash for controversial AI school policy https://www.cbsnews.com/news/ed-markey-google-ai-search-schools/ Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMXTo ensure that you never miss an episode of Transform NOW, be sure to subscribe!

AI-first isn't about automating what exists, it's about asking what should exist. In this episode, host Michael Marchuk welcomes Elisabeth Goos, who leads Roland Berger's AI-first organization and process redesign work, to explore why bolting AI onto legacy processes disappoints, and what radical redesign actually delivers.Key topics include: Why 95% of organizations are stuck in pilot mode despite massive AI investment Zero-based design: how one OEM achieved 30-40% lead time reductions by eliminating entire process steps built only for human handoffs Keeping humans where they add value, and removing them where they're just administrative burden Why "this is how we've always done it" is the hardest obstacle to overcome Building diverse redesign teams that balance deep experience with zero-based thinking How governance becomes an accelerator rather than a wet blanket Why traditional companies have just 6-18 months to move, and what happens if they wait The 40-60% improvement potential across time, quality, and cost vs. incremental optimization Combining radical AI-first redesign with quick wins for a pragmatic path forwardIf your organization is proud of its AI experiments but seeing no ROI, this conversation explains why and what to do about it.Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMXTo ensure that you never miss an episode of Transform NOW, be sure to subscribe!

In this episode, Brad Hairston talks with Dr. Rumman Chowdhury, CEO and co-founder of Humane Intelligence, a tech nonprofit building a community of practice around AI accountability. Dr. Chowdhury explains why "rogue AI" is really a governance problem and how independent audits, red teaming, and clear liability frameworks can make AI trustworthy..What we talked about with Rumman:* Building a professional ecosystem for independent AI auditing and red teaming* Revised Fed model risk guidance, and why trust may beat capability in the AI race* The EU AI Act’s early enforcement: conformity assessments, transparency, and agency* Governing agentic AI through principal-agent theory, permissioning, and authentication* Legal accountability for autonomous agents and the risks of AI personhood* AI sovereignty as an umbrella term, and the closed doors it may create* Why technologists alone shouldn’t decide how AI shapes society..Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMX..To ensure that you never miss an episode of Transform NOW, be sure to subscribe!

In this episode, Transform NOW co-hosts Brad Hairston and Michael Marchuk banter about some interesting recent news headlines in the world of automation and AI. .News stories covered in this episode:.Sam Altman thinks it's 'quite terrible' for your PC to be using memory that AI companies need so badly https://tech.yahoo.com/ai/article/sam-altman-thinks-its-quite-terrible-for-your-pc-to-be-using-memory-that-ai-companies-need-so-badly-165143650.html .Google’s first data center satellite is online and writing poetry https://techcrunch.com/2026/10/30/googles-first-data-center-satellite-is-online-and-writing-poetry/ .AI adoption in 2026: use, spending, and returns https://www.stateof.ai/ai-adoption .DoorDash said it’s only received ‘several hundred’ orders from AI agents https://techcrunch.com/2026/10/30/doordash-said-its-only-received-several-hundred-orders-from-ai-agents .3 Questions: What is the best path forward for AI in academia? https://news.mit.edu/2026/3-questions-best-path-forward-for-ai-academia-1008 .Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMX.To ensure that you never miss an episode of Transform NOW, be sure to subscribe!

In this episode, Brad Hairston talks with Kathleen Walch, an AI strategist and frequent author/speaker at the Project Management Institute, who explains how the CPMAI methodology helps organizations avoid the most common reasons AI projects fail: skipping the business problem, neglecting data, and confusing automation with autonomy..What we discussed with Kathleen:* The origin of CPMAI, from Cognilytica’s 2018 work to PMI’s fastest-growing certification* CPMAI’s six phases, and why teams that skip straight to model development (or building agents) are set up to fail* Overpromising and under-delivering, “pseudo-AI,” and why AI is the “oldest newest technology”* AI literacy as a reflex: perishable, durable and enduring skills* Agentic AI versus automation, and why not every organization needs to build its own agents* AI governance as an organization-wide responsibility, not just IT’s.Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMX.To ensure that you never miss an episode of Transform NOW, be sure to subscribe!

In this episode, host Michael Marchuk talks with Stephan Keese, Managing Partner at Roland Berger North America, on how companies pursuing isolated use cases instead of rebuilding entire processes and functions will likely not see the results they anticipated. Stephan, who leads Roland Berger's industrial operations practice, cuts through the AI hype and discusses what it actually takes for manufacturers to scale.Key topics include:- Why "AI islands" of successful pilots never add up to transformation- How to measure AI ROI beyond headcount reduction (think OEE, quality, cycle time)- Taming messy, siloed shop floor data without starting from scratch- Building operator trust in AI-driven insights- A real-world AI use case that turned 500-page RFQs into one-day responses- Why composable, AI-ready architecture beats off-the-shelf solutions- The future of humanoid robots, re-shoring, and the social challenges aheadWhether you're a manufacturer exploring your first use case or a leader pushing past pilot purgatory, this conversation offers a practical roadmap for embedding AI into your operating model before your competitors do.Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMXTo ensure that you never miss an episode of Transform NOW, be sure to subscribe!

In this episode, Transform NOW co-hosts Brad Hairston and Michael Marchuk banter about some interesting recent news headlines in the world of automation and AI. News stories covered in this episode:• Anthropic releases IPO filing ahead of highly-anticipated stock debut https://www.nbcnews.com/business/markets/anthropic-releases-ipo-filing-ahead-highly-anticipated-stock-debut-rcna605676 • AI governance gaps leave RIAs exposed to a new wave of cyber threats https://www.investmentnews.com/practice-management/ai-governance-gaps-leave-rias-exposed-to-a-new-wave-of-cyber-threats/268735 • Open models gain traction in enterprises, tech CEOs say - https://www.constellationr.com/insights/news/open-models-gain-traction-enterprises-tech-ceos-say • Renting the frontier, owning the core IP - https://www.thoughtworks.com/en-us/insights/blog/generative-ai/renting-the-frontier--owning-the-core-ip--how-we-cut-ai-forecast • The molecule came from AI. The inventor still has to be human https://cen.acs.org/acs-news/comment-molecule-came-ai-inventor/104/web/2026/10 Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMX.To ensure that you never miss an episode of Transform NOW, be sure to subscribe!

AI agents will become the next mushroom cloud in organizations and most companies aren't ready. Denis O'Shea, founder of Mobile Mentor, joins the Transform Now podcast to share zero trust architecture lessons for the agentic era, starting with his own company's honest mistakes.In this episode:- Mobile Mentor's ugly story: 33,000 over-shared data assets and no ROI measurement- The five workstreams that fix it: use cases, data security, user empowerment, agent frameworks, ROI- Why personal agents are harmless but orphaned enterprise agents are a ticking time bomb- The "Janet problem": when the agent's creator leaves and nobody knows how it works- Why enterprise agents need service accounts, DevOps discipline, and pro development- Gartner's prediction: 100 agents per employee and why we need an agent control plane- Data flows: the next frontier of data security posture management- The cost problem: credits, tokens, overages, and deciphering vendor billing- Where AI ownership should sit (hint: not with the CIO alone)• Mobile Mentor website: https://www.mobile-mentor.com/• Denis’s LinkedIn: https://www.linkedin.com/in/denisosheamobilementor/• Mobile Mentor’s LinkedIn: https://www.linkedin.com/company/mobile-mentor-ltd/• Additional social handles: https://www.denisoshea.com/Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMXTo ensure that you never miss an episode of Transform NOW, be sure to subscribe!

In this episode, Brad Hairston talks with Michael Fauscette, founder and chief analyst at Arion Research, who explains why most agentic AI pilots never make it to production, and how companies can get there by giving agents goals instead of scripts, redesigning workflows, and building governance in from the start..What we talked about with Michael:* A continuum of autonomy - what separates agentic AI from deterministic automation. An agent becomes a “coworker” when it can take a goal instead of a script, then plan, act and learn.*The gap between pilots and production - many companies experiment, but few scale. They get stuck on integration, data, skills, cost, compliance and change management.* Workflow redesign over agent drop-ins - humans quietly fix broken process steps that agents will repeat and compound at machine speed.* Error tolerance by use case - a 5% error rate is a UX annoyance in a chatbot, but a compliance and financial exposure when an agent takes actions.* The risk of a badly set goal - “keep the customer happy” can turn into 100 unauthorized refunds.* Total cost of ownership at scale - a low-volume pilot’s costs tell you little about the production cost curve, and hidden change management and token costs pile up.* Governance and orchestration designed in from the start - boundary conditions, plus human-in-the-loop, human-on-the-loop and human-in-the-lead oversight models.* A metric for scaling - each new use case should get cheaper and faster to stand up than the last..Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMX..To ensure that you never miss an episode of Transform NOW, be sure to subscribe!

In this episode, Transform NOW co-hosts Brad Hairston and Michael Marchuk banter about some interesting recent news headlines in the world of automation and AI. .News stories covered in this episode:.AI model decodes the language cells use to communicate https://news.mit.edu/2026/ai-model-decodes-language-cells-use-communicate-1015.SpaceX And Tesla Are Building A Massive New Factory, And Here's What It'll Make https://finance.yahoo.com/technology/articles/spacex-tesla-building-massive-factory-022500027.html .Anthropic warns of imminent self-replicating AI that might even work off an oven or TV https://www.yahoo.com/tech/ai/claude/articles/anthropic-warns-imminent-self-replicating-182122260.html Research discussed in the article: https://www.anthropic.com/research/self-replicating-ai.Pulling the plug: Why the AI kill switch might be a dead end https://www.techtarget.com/cybersecurity/news/366651197/Pulling-the-plug-Why-the-AI-kill-switch-might-be-a-dead-end .Even LinkedIn is getting tired of your AI-written posts https://www.cnn.com/2026/10/18/tech/linkedin-fighting-ai-p .Visit us on our socials:🦾 Get started with SS&C Blue Prism: https://okt.to/JcMLdU🧑‍💻LinkedIn: https://okt.to/k8zIdp✖️Twitter: https://okt.to/fHyd9G🙋‍♀️Facebook: https://okt.to/Vyjfiz📸Instagram: https://okt.to/5nYvIf💭Blog: https://okt.to/QuGqVP🤩Case studies: https://okt.to/ft1AMX.To ensure that you never miss an episode of Transform NOW, be sure to subscribe!