[
  {
    "start": 0.06,
    "end": 8.3,
    "text": "This episode is provided by Thomas Allgaier and Frenes based on the most relevant LinkedIn posts about artificial intelligence in calendar weeks, forty one and forty two."
  },
  {
    "start": 9.54,
    "end": 18.44,
    "text": "Frenes supports enterprises with market and competitive intelligence, decoding emerging technologies, customer insights, regulatory shifts and competitor strategies."
  },
  {
    "start": 18.9,
    "end": 23.98,
    "text": "So product teams and strategy leaders don't just react, but shape the future of AI."
  },
  {
    "start": 24.94,
    "end": 25.44,
    "text": "Welcome back."
  },
  {
    "start": 26.14,
    "end": 34.08,
    "text": "So today we're really going to dig into the top artificial intelligence trends we've been seeing across professional insights from the last couple of weeks."
  },
  {
    "start": 34.24,
    "end": 34.44,
    "text": "Right."
  },
  {
    "start": 34.8,
    "end": 43.5,
    "text": "The mission really is to figure out where things are shifting, you know, moving past the hype, the pilots towards disciplined, measurable execution, real enterprise stuff."
  },
  {
    "start": 43.66,
    "end": 44.08,
    "text": "Exactly."
  },
  {
    "start": 44.2,
    "end": 45.26,
    "text": "Execution, that's the word."
  },
  {
    "start": 45.3,
    "end": 45.46,
    "text": "Yeah."
  },
  {
    "start": 45.62,
    "end": 47.32,
    "text": "The conversation feels like it's grown up a bit."
  },
  {
    "start": 48.44,
    "end": 51.5,
    "text": "We've kind of clustered the insights for you today into three main areas."
  },
  {
    "start": 51.6,
    "end": 51.82,
    "text": "Okay."
  },
  {
    "start": 52.04,
    "end": 59.6,
    "text": "We'll kick off with the foundation infrastructure, the chip wars, up and how these new strategic motes are being dug."
  },
  {
    "start": 60.28,
    "end": 66.2,
    "text": "Then we'll jump into the application layer, particularly a genetic AI and this whole workflow revolution people are talking about."
  },
  {
    "start": 66.26,
    "end": 67.14,
    "text": "Yeah, that's a big one."
  },
  {
    "start": 67.34,
    "end": 71.6,
    "text": "And finally, we have to tackle the non-technical side, which is often the hardest part."
  },
  {
    "start": 72.26,
    "end": 76.48,
    "text": "Culture, risk, and how to adopt this responsibly."
  },
  {
    "start": 76.74,
    "end": 77.12,
    "text": "Good plan."
  },
  {
    "start": 77.82,
    "end": 79.46,
    "text": "So let's unpack that foundation first."
  },
  {
    "start": 79.84,
    "end": 84.14,
    "text": "What's really fascinating, I think, is the high stakes game being played around compute power."
  },
  {
    "start": 84.4,
    "end": 84.74,
    "text": "Definitely."
  },
  {
    "start": 84.88,
    "end": 88.24,
    "text": "The whole idea of an AI moat is absolutely shifting."
  },
  {
    "start": 88.34,
    "end": 90.74,
    "text": "It's not just about having the best model anymore."
  },
  {
    "start": 90.84,
    "end": 94.1,
    "text": "It's becoming about who owns the entire pipeline."
  },
  {
    "start": 94.16,
    "end": 94.78,
    "text": "The whole stack."
  },
  {
    "start": 94.82,
    "end": 96.52,
    "text": "The whole vertical compute pipeline, yeah."
  },
  {
    "start": 97.08,
    "end": 98.66,
    "text": "From the silicon right up to the user."
  },
  {
    "start": 98.92,
    "end": 106.08,
    "text": "And maybe the biggest sign of this huge shift is seeing NVIDIA's grip on the GPU market starts to loosen a bit."
  },
  {
    "start": 106.24,
    "end": 106.5,
    "text": "Right."
  },
  {
    "start": 106.68,
    "end": 109.22,
    "text": "The standard for AI infrastructure for so long."
  },
  {
    "start": 109.38,
    "end": 109.76,
    "text": "Exactly."
  },
  {
    "start": 110.18,
    "end": 112.16,
    "text": "And Nehanth Maidu-Kalasetti pointed this out."
  },
  {
    "start": 112.5,
    "end": 113.86,
    "text": "Oracle confirmed they're deploying what?"
  },
  {
    "start": 114.0,
    "end": 116.28,
    "text": "Fifty thousand AMD AI chips?"
  },
  {
    "start": 116.58,
    "end": 117.54,
    "text": "The MI-Four fifties?"
  },
  {
    "start": 117.86,
    "end": 119.14,
    "text": "Starting mid-twenty-twenty-six."
  },
  {
    "start": 119.66,
    "end": 120.64,
    "text": "Fifty thousand?"
  },
  {
    "start": 121.2,
    "end": 123.26,
    "text": "That is a, well, a staggering number."
  },
  {
    "start": 124.3,
    "end": 126.58,
    "text": "And it's not just about having a second source, right?"
  },
  {
    "start": 126.7,
    "end": 129.479,
    "text": "This looks like a really major alliance taking shape."
  },
  {
    "start": 129.699,
    "end": 130.36,
    "text": "Oracle, AMD."
  },
  {
    "start": 130.66,
    "end": 130.8,
    "text": "And"
  },
  {
    "start": 130.88,
    "end": 131.42,
    "text": "OpenAI."
  },
  {
    "start": 131.82,
    "end": 139.4,
    "text": "They're making a huge bet that owning and optimizing these specific compute pipelines, well, that's a new oil field, basically."
  },
  {
    "start": 139.5,
    "end": 139.56,
    "text": "It"
  },
  {
    "start": 139.62,
    "end": 140.16,
    "text": "feels like it."
  },
  {
    "start": 140.22,
    "end": 145.64,
    "text": "I mean, that Oracle Cloud deal with OpenAI alone, potentially worth up to three hundred billion dollars."
  },
  {
    "start": 145.68,
    "end": 145.8,
    "text": "Three"
  },
  {
    "start": 145.86,
    "end": 146.84,
    "text": "hundred billion, yeah."
  },
  {
    "start": 147.28,
    "end": 149.66,
    "text": "That valuation tells you everything."
  },
  {
    "start": 150.04,
    "end": 156.0,
    "text": "about the race cloud providers are in to lockdown supply and, you know, integrate hardware design right into their services."
  },
  {
    "start": 156.1,
    "end": 157.58,
    "text": "It really is a full stack war now."
  },
  {
    "start": 158.16,
    "end": 161.32,
    "text": "Makes it much tougher for newcomers trying to compete just on the model itself."
  },
  {
    "start": 161.42,
    "end": 161.52,
    "text": "But,"
  },
  {
    "start": 161.7,
    "end": 166.44,
    "text": "okay, even if you figure out the chip supply, you immediately bump up against some pretty fundamental limits."
  },
  {
    "start": 166.6,
    "end": 169.24,
    "text": "Walter Zuzavi highlighted this for major economies."
  },
  {
    "start": 169.44,
    "end": 169.58,
    "text": "Like"
  },
  {
    "start": 169.62,
    "end": 170.48,
    "text": "the physical barriers."
  },
  {
    "start": 170.7,
    "end": 171.08,
    "text": "Exactly."
  },
  {
    "start": 171.38,
    "end": 175.3,
    "text": "In places like Germany, for instance, AI ambition is hitting a, well, a very hard wall."
  },
  {
    "start": 176.5,
    "end": 178.88,
    "text": "Just basic data center power and capacity and"
  },
  {
    "start": 178.96,
    "end": 183.94,
    "text": "I saw those numbers the planned it load for twenty thirty something like three point seven gigawatts."
  },
  {
    "start": 184.36,
    "end": 189.0,
    "text": "Yeah, but the projected demand could be closer to like, five point nine gigawatts."
  },
  {
    "start": 189.06,
    "end": 189.24,
    "text": "Wow."
  },
  {
    "start": 189.48,
    "end": 191.44,
    "text": "That gap was at two point two gigawatts."
  },
  {
    "start": 191.5,
    "end": 192.88,
    "text": "That's a massive shortfall."
  },
  {
    "start": 193.42,
    "end": 200.02,
    "text": "It definitely shifts capacity planning from just operations to a serious strategic risk at the board level."
  },
  {
    "start": 200.14,
    "end": 200.64,
    "text": "Absolutely."
  },
  {
    "start": 200.72,
    "end": 210.68,
    "text": "You can't really chase ambitious AI strategies, especially regulated ones, or build out sovereign clouds if you literally can't guarantee the power to run the hardware."
  },
  {
    "start": 210.9,
    "end": 213.02,
    "text": "And that leads right into the next bottleneck, doesn't it?"
  },
  {
    "start": 213.62,
    "end": 214.18,
    "text": "connectivity."
  },
  {
    "start": 214.34,
    "end": 214.84,
    "text": "It does."
  },
  {
    "start": 215.64,
    "end": 222.4,
    "text": "Alan Weckel noted that these huge AI networks often stumble not because the GPUs aren't fast enough, but because of the underlying network fabric."
  },
  {
    "start": 222.64,
    "end": 223.62,
    "text": "Ah, the fabric."
  },
  {
    "start": 223.84,
    "end": 230.08,
    "text": "So for people maybe less deep in networking, we're talking the high-speed connections linking all those thousands of GPUs."
  },
  {
    "start": 230.2,
    "end": 230.7,
    "text": "Precisely."
  },
  {
    "start": 231.1,
    "end": 232.9,
    "text": "Option non-blocking Ethernet designs."
  },
  {
    "start": 233.54,
    "end": 239.26,
    "text": "If that fabric isn't absolutely perfect, the second you try to distribute a really large training job, performance just tanks."
  },
  {
    "start": 239.34,
    "end": 239.52,
    "text": "Right."
  },
  {
    "start": 239.62,
    "end": 240.44,
    "text": "It grinds to a halt."
  },
  {
    "start": 240.78,
    "end": 240.92,
    "text": "Yeah."
  },
  {
    "start": 241.6,
    "end": 243.92,
    "text": "The fabric is like the central nervous system."
  },
  {
    "start": 244.12,
    "end": 248.38,
    "text": "It has to handle enormous amounts of data, all moving in parallel, no collisions."
  },
  {
    "start": 249.06,
    "end": 256.04,
    "text": "These non-blocking designs are basically non-negotiable now for the kind of scale you need in, say, Neo or sovereign clouds."
  },
  {
    "start": 256.3,
    "end": 256.66,
    "text": "Thanks, us."
  },
  {
    "start": 256.839,
    "end": 263.0,
    "text": "And Rick Moore mentioned solutions like Lumen's Wavelength Rapid Routes offering, you know, four hundred GBPs dedicated links."
  },
  {
    "start": 263.58,
    "end": 268.74,
    "text": "They're specifically built to kill those bottlenecks, especially when your architecture is spread out geographically."
  },
  {
    "start": 268.98,
    "end": 276.7,
    "text": "Okay, so... That focus on connectivity, performance at scale, it ties directly into what Armand Ruiz observed, doesn't it?"
  },
  {
    "start": 277.42,
    "end": 281.4,
    "text": "This shift in industry focus away from the super expensive training phase."
  },
  {
    "start": 281.54,
    "end": 281.74,
    "text": "Torch"
  },
  {
    "start": 281.84,
    "end": 282.54,
    "text": "inference, yeah."
  },
  {
    "start": 282.66,
    "end": 283.06,
    "text": "Inference."
  },
  {
    "start": 283.36,
    "end": 284.7,
    "text": "Model deployment, execution."
  },
  {
    "start": 285.1,
    "end": 287.18,
    "text": "That's what dominates the long-term AI workload."
  },
  {
    "start": 287.38,
    "end": 292.5,
    "text": "Training is that big upfront cost, sure, but... inference is the ongoing operational cost."
  },
  {
    "start": 292.64,
    "end": 298.12,
    "text": "So the drive for efficiency there, especially at the edge where the data often is, that's where the real ROI is going to be."
  },
  {
    "start": 298.2,
    "end": 298.38,
    "text": "That's"
  },
  {
    "start": 298.46,
    "end": 298.96,
    "text": "the thinking."
  },
  {
    "start": 299.14,
    "end": 301.88,
    "text": "It's all about efficient execution for lasting value."
  },
  {
    "start": 302.32,
    "end": 309.88,
    "text": "I do wonder though, if the focus shifts so hard to inference efficiency, does that mean training budgets actually shrink eventually?"
  },
  {
    "start": 310.26,
    "end": 312.98,
    "text": "Or does the complexity just keep pushing the training costs up anyway?"
  },
  {
    "start": 313.2,
    "end": 314.74,
    "text": "That's the million dollar question, isn't it?"
  },
  {
    "start": 314.84,
    "end": 316.2,
    "text": "Or maybe the billion dollar question now."
  },
  {
    "start": 316.3,
    "end": 320.38,
    "text": "But the immediate goal is definitely ROI, getting value out of these things."
  },
  {
    "start": 320.42,
    "end": 323.92,
    "text": "And that sets us up perfectly for our next theme, the agentic enterprise."
  },
  {
    "start": 324.06,
    "end": 324.28,
    "text": "Right."
  },
  {
    "start": 324.8,
    "end": 334.48,
    "text": "Because that need for efficient inference is exactly what fuels the drive for these more autonomous systems, systems that can actually handle multi-step workflows."
  },
  {
    "start": 334.6,
    "end": 336.06,
    "text": "We are definitely seeing that shift."
  },
  {
    "start": 336.46,
    "end": 347.16,
    "text": "Moving beyond just static tools to systems that can, you know, reason, plan, act on their own, Tony Moroni laid out four ways these agents could unlock serious enterprise value."
  },
  {
    "start": 347.42,
    "end": 348.0,
    "text": "Okay, like what?"
  },
  {
    "start": 348.24,
    "end": 349.82,
    "text": "Well, expanding automation, obviously."
  },
  {
    "start": 350.14,
    "end": 351.98,
    "text": "Having systems available to it over for four or seven."
  },
  {
    "start": 352.78,
    "end": 360.68,
    "text": "building more organizational resilience, and, maybe most importantly, turning all the dormant knowledge companies have directly into action."
  },
  {
    "start": 361.1,
    "end": 361.64,
    "text": "Verifiable"
  },
  {
    "start": 361.74,
    "end": 361.96,
    "text": "action."
  },
  {
    "start": 362.26,
    "end": 365.08,
    "text": "And to do that, you need the right technical underpinning."
  },
  {
    "start": 365.46,
    "end": 369.54,
    "text": "That's why Eduardo Ordex called agentic rag plus memory the missing link."
  },
  {
    "start": 370.06,
    "end": 373.22,
    "text": "OK, so rag, retrieval augmented generation, most people know that."
  },
  {
    "start": 373.26,
    "end": 375.42,
    "text": "It's how you ground the model in your own documents, right?"
  },
  {
    "start": 375.66,
    "end": 375.72,
    "text": "Yeah."
  },
  {
    "start": 375.76,
    "end": 376.74,
    "text": "Stop from just making things."
  },
  {
    "start": 376.8,
    "end": 377.18,
    "text": "Exactly."
  },
  {
    "start": 377.26,
    "end": 378.98,
    "text": "Stops the hallucinations or tries to."
  },
  {
    "start": 379.74,
    "end": 384.96,
    "text": "But agentic rag plus memory adds three really key things on top."
  },
  {
    "start": 385.74,
    "end": 390.6,
    "text": "An orchestrator, specific memory functions, and these reasoning loops, like react."
  },
  {
    "start": 390.9,
    "end": 391.02,
    "text": "OK."
  },
  {
    "start": 391.38,
    "end": 393.86,
    "text": "which lets the system actually plan dynamically."
  },
  {
    "start": 394.32,
    "end": 400.52,
    "text": "Not just grab a chunk of text, but break down a complex task, search iteratively, synthesize info over multiple steps."
  },
  {
    "start": 400.88,
    "end": 403.24,
    "text": "You know, solve real messy enterprise problems."
  },
  {
    "start": 403.32,
    "end": 414.02,
    "text": "So instead of just asking what was Q three revenue and getting a number, you could analyze Q three revenue against Q four cost projections and suggest three ways to cut costs."
  },
  {
    "start": 414.88,
    "end": 415.78,
    "text": "Something that needs planning."
  },
  {
    "start": 416.06,
    "end": 416.8,
    "text": "That's the idea."
  },
  {
    "start": 416.94,
    "end": 418.24,
    "text": "That's the practical leap forward."
  },
  {
    "start": 418.28,
    "end": 420.2,
    "text": "And we're seeing this pop up in actual commerce."
  },
  {
    "start": 420.38,
    "end": 426.02,
    "text": "Hugo Caterino mentioned how in agentic search, these assistants are trying to unify discovery, trust, and even settlement."
  },
  {
    "start": 426.22,
    "end": 432.0,
    "text": "So search relevance suddenly has to include checking price, stock, whether the payment method works."
  },
  {
    "start": 433.44,
    "end": 437.22,
    "text": "because if the agent suggests buying it, it better be immediately doable."
  },
  {
    "start": 437.34,
    "end": 439.24,
    "text": "That really raises the stakes, doesn't it?"
  },
  {
    "start": 439.46,
    "end": 442.56,
    "text": "The liability and the expectations on the AI in a transaction?"
  },
  {
    "start": 442.88,
    "end": 445.46,
    "text": "It absolutely does, but it also creates opportunities."
  },
  {
    "start": 445.56,
    "end": 450.38,
    "text": "Anna Lerner-Nesbitt was talking about the Walmart and open AI shopping experience."
  },
  {
    "start": 450.44,
    "end": 450.7,
    "text": "Oh, yes,"
  },
  {
    "start": 450.78,
    "end": 450.84,
    "text": "all"
  },
  {
    "start": 450.88,
    "end": 451.02,
    "text": "that."
  },
  {
    "start": 451.06,
    "end": 452.8,
    "text": "And she raised a really interesting point."
  },
  {
    "start": 453.44,
    "end": 458.74,
    "text": "Agents could be a huge opportunity to bake in sustainable choices from the start."
  },
  {
    "start": 458.9,
    "end": 459.14,
    "text": "How so?"
  },
  {
    "start": 459.9,
    "end": 462.88,
    "text": "Well, they could act like automated choice architects, right?"
  },
  {
    "start": 463.42,
    "end": 474.08,
    "text": "Prioritizing efficiency, maybe suggesting lower waste options, guiding users towards more sustainable choices across the value chain, potentially without needing constant human intervention."
  },
  {
    "start": 474.16,
    "end": 475.34,
    "text": "That's a powerful concept."
  },
  {
    "start": 475.72,
    "end": 478.96,
    "text": "Using automation to nudge behavior towards positive outcomes."
  },
  {
    "start": 479.82,
    "end": 481.4,
    "text": "But we have to get real about deployment."
  },
  {
    "start": 481.48,
    "end": 482.48,
    "text": "It's proving really tough."
  },
  {
    "start": 482.6,
    "end": 483.4,
    "text": "Oh, absolutely."
  },
  {
    "start": 484.0,
    "end": 490.16,
    "text": "Virtrungland.DA cited an HBR report pointing out that most companies are really struggling to scale AI beyond the pussies"
  },
  {
    "start": 490.24,
    "end": 490.52,
    "text": "stage."
  },
  {
    "start": 490.76,
    "end": 491.18,
    "text": "Why is that?"
  },
  {
    "start": 491.44,
    "end": 496.06,
    "text": "Performance drops, unpredictable costs, and big security gaps are the main culprits."
  },
  {
    "start": 496.16,
    "end": 497.06,
    "text": "Okay, so what's the fix?"
  },
  {
    "start": 497.52,
    "end": 502.18,
    "text": "The emerging consensus seems to be around something called generative operations, or GenOps."
  },
  {
    "start": 502.96,
    "end": 507.24,
    "text": "Think MLO's rigor, but specifically tailored for these generative models."
  },
  {
    "start": 507.48,
    "end": 509.26,
    "text": "So a structured way to manage deployment."
  },
  {
    "start": 509.46,
    "end": 510.0,
    "text": "Exactly."
  },
  {
    "start": 510.28,
    "end": 516.28,
    "text": "Automated monitoring, drift detection, cost analysis, governance checks baked in before, during, and after deployment."
  },
  {
    "start": 516.84,
    "end": 522.22,
    "text": "Without that kind of operational discipline, these agents just won't scale responsibly or maybe even safely."
  },
  {
    "start": 522.659,
    "end": 524.14,
    "text": "Which brings us right to our last theme."
  },
  {
    "start": 524.82,
    "end": 527.38,
    "text": "Culture, risk, and responsible adoption."
  },
  {
    "start": 527.54,
    "end": 527.7,
    "text": "Yeah."
  },
  {
    "start": 528.22,
    "end": 534.5,
    "text": "Because the best infrastructure and the smartest agents, well, they're kind of useless if the organization isn't ready."
  },
  {
    "start": 534.88,
    "end": 536.28,
    "text": "That really is the heart of it, isn't it?"
  },
  {
    "start": 536.36,
    "end": 538.62,
    "text": "The biggest hurdle often isn't technical at all."
  },
  {
    "start": 538.78,
    "end": 540.74,
    "text": "There is an IBM survey of execs."
  },
  {
    "start": 541.04,
    "end": 541.2,
    "text": "Yeah."
  },
  {
    "start": 541.48,
    "end": 545.42,
    "text": "Sixty-seven percent expect agents to act independently by twenty-twenty-seven."
  },
  {
    "start": 545.94,
    "end": 546.6,
    "text": "Huge number."
  },
  {
    "start": 547.18,
    "end": 550.06,
    "text": "But the main barrier they see, not the tech, it's"
  },
  {
    "start": 550.16,
    "end": 550.52,
    "text": "people,"
  },
  {
    "start": 551.02,
    "end": 553.04,
    "text": "mindset, organizational readiness."
  },
  {
    "start": 553.32,
    "end": 554.34,
    "text": "It always comes back to the people."
  },
  {
    "start": 555.24,
    "end": 557.68,
    "text": "Pretty, WaterD made that point really clearly."
  },
  {
    "start": 558.34,
    "end": 562.42,
    "text": "Sustainable AI impact comes from prepared people, not just powerful platforms."
  },
  {
    "start": 562.54,
    "end": 562.74,
    "text": "Right."
  },
  {
    "start": 563.02,
    "end": 563.9,
    "text": "And the stats back it up."
  },
  {
    "start": 564.36,
    "end": 572.58,
    "text": "Something like seventy-eight percent of global orgs are using AI now, but fewer than one percent are actually considered fully mature in how they use it."
  },
  {
    "start": 572.78,
    "end": 575.46,
    "text": "And you see that maturity gap reflected nationally too."
  },
  {
    "start": 575.86,
    "end": 579.28,
    "text": "Donor Mustafa highlighted the situation in Sweden pretty striking."
  },
  {
    "start": 579.92,
    "end": 586.38,
    "text": "Seventy-seven percent of Swedes using AI at work have had zero formal education or training on it."
  },
  {
    "start": 586.78,
    "end": 587.62,
    "text": "seventy seven percent."
  },
  {
    "start": 587.9,
    "end": 588.7,
    "text": "wow yeah that's."
  },
  {
    "start": 589.26,
    "end": 591.58,
    "text": "that's not an employee problem that's a leadership failure."
  },
  {
    "start": 591.68,
    "end": 592.58,
    "text": "surely absolutely."
  },
  {
    "start": 592.66,
    "end": 594.6,
    "text": "when adoption happens bottom up like that."
  },
  {
    "start": 594.74,
    "end": 599.24,
    "text": "and leadership doesn't provide guardrails doesn't provide training that's a massive risk"
  },
  {
    "start": 599.44,
    "end": 606.94,
    "text": "which ties into navine balani's reminder that ai really needs to be treated as an engineering discipline not just you know a bunch of cool experiments"
  },
  {
    "start": 607.12,
    "end": 607.46,
    "text": "totally."
  },
  {
    "start": 607.8,
    "end": 610.8,
    "text": "The differentiator isn't just having the latest LLM or a co-pilot."
  },
  {
    "start": 611.26,
    "end": 616.7,
    "text": "It's the engineering mindset behind how you deploy, manage, and integrate these tools into critical workflows."
  },
  {
    "start": 617.2,
    "end": 618.06,
    "text": "Rigor matters."
  },
  {
    "start": 618.48,
    "end": 621.02,
    "text": "And that rigor needs to extend down to the end user too."
  },
  {
    "start": 622.28,
    "end": 627.36,
    "text": "Daniel Paul has some really practical advice on getting more out of existing models, like GPT-V."
  },
  {
    "start": 627.6,
    "end": 627.76,
    "text": "Yeah."
  },
  {
    "start": 628.44,
    "end": 629.96,
    "text": "Using its different modes properly."
  },
  {
    "start": 630.26,
    "end": 640.34,
    "text": "like instant mode for quick ideas, thinking mode for strategy, pro mode for deep, precise work, auto mode, if you're not sure, moving beyond just tossing random prompts at it."
  },
  {
    "start": 640.44,
    "end": 640.54,
    "text": "That"
  },
  {
    "start": 640.6,
    "end": 641.02,
    "text": "makes sense."
  },
  {
    "start": 641.08,
    "end": 648.1,
    "text": "Getting intentional about how you interact with it, because that attention to detail is critical, especially when you factor in the ethical stuff and the risks involved."
  },
  {
    "start": 648.16,
    "end": 648.5,
    "text": "Definitely."
  },
  {
    "start": 648.7,
    "end": 651.56,
    "text": "And one of the most sobering things we saw came from Dr."
  },
  {
    "start": 651.6,
    "end": 653.7,
    "text": "Hans Ruznick citing that Harvard study."
  },
  {
    "start": 653.9,
    "end": 653.96,
    "text": "The"
  },
  {
    "start": 654.12,
    "end": 655.14,
    "text": "weird bias one."
  },
  {
    "start": 655.22,
    "end": 655.62,
    "text": "Exactly."
  },
  {
    "start": 655.72,
    "end": 662.06,
    "text": "Showing how LLMs are heavily biased towards Western educated industrial rich and democratic perspectives."
  },
  {
    "start": 662.46,
    "end": 663.76,
    "text": "That's a fundamental challenge."
  },
  {
    "start": 663.86,
    "end": 670.66,
    "text": "If the training data is skewed towards one narrow cultural viewpoint, how fair or applicable are these models globally?"
  },
  {
    "start": 671.1,
    "end": 677.0,
    "text": "It raises huge questions about governance, trust, especially for decisions impacting devotes groups."
  },
  {
    "start": 677.16,
    "end": 677.52,
    "text": "For sure."
  },
  {
    "start": 677.88,
    "end": 682.8,
    "text": "And even when the output sounds right, Kassi Kazarkov warns about the fluency heuristic."
  },
  {
    "start": 683.18,
    "end": 686.78,
    "text": "Ah, yeah, the tendency to trust something just because it sounds smooth and confident."
  },
  {
    "start": 686.9,
    "end": 687.32,
    "text": "Exactly."
  },
  {
    "start": 687.46,
    "end": 689.22,
    "text": "It's a dangerous psychological trap."
  },
  {
    "start": 689.84,
    "end": 696.04,
    "text": "The AI sounds so sure of itself that humans might skip the crucial step of actually verifying the output."
  },
  {
    "start": 696.18,
    "end": 698.1,
    "text": "Which can lead to really costly mistakes, right?"
  },
  {
    "start": 698.14,
    "end": 701.4,
    "text": "Because the model will hallucinate or make subtle errors eventually."
  },
  {
    "start": 701.44,
    "end": 705.02,
    "text": "You almost have to default to assuming it's wrong until you prove it right."
  },
  {
    "start": 705.32,
    "end": 706.82,
    "text": "And the final piece of the risk puzzle."
  },
  {
    "start": 707.86,
    "end": 708.32,
    "text": "Security."
  },
  {
    "start": 708.76,
    "end": 710.12,
    "text": "with these autonomous agents."
  },
  {
    "start": 710.92,
    "end": 716.06,
    "text": "Hussaini Savawala pointed out that AI coding agents, well, they're not just speeding up development."
  },
  {
    "start": 716.1,
    "end": 718.38,
    "text": "They're speeding up vulnerability creation, too."
  },
  {
    "start": 718.52,
    "end": 719.18,
    "text": "Unfortunately, yeah."
  },
  {
    "start": 719.6,
    "end": 723.14,
    "text": "Replicating security flaws at enterprise scale incredibly quickly."
  },
  {
    "start": 724.48,
    "end": 725.7,
    "text": "So what's the answer there?"
  },
  {
    "start": 726.04,
    "end": 728.82,
    "text": "It seems to be what they call data-first agentic security."
  },
  {
    "start": 729.48,
    "end": 731.54,
    "text": "Basically, build security in from the start."
  },
  {
    "start": 732.1,
    "end": 736.12,
    "text": "prevent sensitive data from even getting into the model's context if possible."
  },
  {
    "start": 736.5,
    "end": 738.2,
    "text": "Don't try to just patch things later."
  },
  {
    "start": 738.48,
    "end": 738.88,
    "text": "Makes sense."
  },
  {
    "start": 739.24,
    "end": 741.7,
    "text": "Security as a foundation, not an afterthought."
  },
  {
    "start": 742.1,
    "end": 745.68,
    "text": "So pulling it all together from the insights these past couple of weeks."
  },
  {
    "start": 745.78,
    "end": 745.88,
    "text": "Yeah."
  },
  {
    "start": 746.48,
    "end": 747.46,
    "text": "What's the big picture?"
  },
  {
    "start": 748.36,
    "end": 752.78,
    "text": "It really feels like the market's moving past the initial hype, getting down to the foundations."
  },
  {
    "start": 752.84,
    "end": 753.92,
    "text": "Definitely maturing."
  },
  {
    "start": 754.1,
    "end": 754.24,
    "text": "Yeah."
  },
  {
    "start": 754.58,
    "end": 761.62,
    "text": "And that demands serious strategic investment in infrastructure, whether it's new chip deals or perfecting that network fabric."
  },
  {
    "start": 762.16,
    "end": 770.8,
    "text": "But maybe even more crucial is the shift needed in company culture and engineering discipline to actually manage these powerful new agentic systems properly."
  },
  {
    "start": 771.08,
    "end": 773.12,
    "text": "And Kira Dotson put it pretty bluntly."
  },
  {
    "start": 773.64,
    "end": 773.76,
    "text": "Yeah."
  },
  {
    "start": 774.1,
    "end": 777.2,
    "text": "The era of AI for the sake of AI is over."
  },
  {
    "start": 777.58,
    "end": 777.76,
    "text": "Right."
  },
  {
    "start": 777.86,
    "end": 784.48,
    "text": "So the question every organization needs to ask itself now is, are we still chasing vanity metrics just looking busy with AI?"
  },
  {
    "start": 784.96,
    "end": 792.96,
    "text": "Or are we truly focused on delivering measurable enterprise grade value that's going to last beyond the current hype cycle and actually reshape how we operate?"
  },
  {
    "start": 793.3,
    "end": 794.84,
    "text": "That really is the key differentiator now."
  },
  {
    "start": 795.02,
    "end": 796.02,
    "text": "Focus on real value."
  },
  {
    "start": 796.14,
    "end": 796.7,
    "text": "Exactly."
  },
  {
    "start": 797.0,
    "end": 800.36,
    "text": "Are you building something that endures or just riding the wave?"
  },
  {
    "start": 800.74,
    "end": 801.18,
    "text": "Well said."
  },
  {
    "start": 801.8,
    "end": 805.2,
    "text": "If you enjoyed this deep dive, new episodes drop every two weeks."
  },
  {
    "start": 805.78,
    "end": 808.06,
    "text": "Also check out our other editions on ICT and tech."
  },
  {
    "start": 808.42,
    "end": 814.54,
    "text": "Digital products and services, cloud, sustainability in green ICT, defense tech and health tech."
  },
  {
    "start": 814.9,
    "end": 815.58,
    "text": "Thanks for tuning in."
  },
  {
    "start": 815.76,
    "end": 821.68,
    "text": "We really encourage you to subscribe so you don't miss our continued deep dives and to the insights that matter most for your strategy."
  }
]