Latest / Best of LinkedIn: Digital Transformation & Tech / Best of LinkedIn: Digital Products & Services CW 02/ 03

Transcript

Best of LinkedIn: Digital Products & Services CW 02/ 03 · Best of LinkedIn: Digital Transformation & Tech

• 0:00
  Speaker 0: This episode is provided by Thomas Allgaier and Frennus based on the most relevant LinkedIn posts about digital products and services in calendar weeks O two and O three.
• 0:10
  Speaker 0: Frennus is a B to B market research company helping enterprises gain the market customer and competitive insights needed to drive customer centric and cost efficient product development.
• 0:20
  Speaker 1: It's great to be back.
• 0:20
  Speaker 1: We're doing a really specific deep dive today looking at digital products and services at the start of twenty twenty six.
• 0:27
  Speaker 0: It really feels like the dust is finally settling.
• 0:30
  Speaker 0: The last couple of years were just this frantic AI gold rush.
• 0:35
  Speaker 0: Everyone was just scrambling to slap a generative sticker on their product.
• 0:38
  Speaker 0: Right,
• 0:38
  Speaker 1: but early twenty-twenty-six feels different.
• 0:41
  Speaker 1: Yeah.
• 0:41
  Speaker 1: We're seeing this pretty clear shift from hype to, well, reliability.
• 0:46
  Speaker 1: The talk isn't about what AI could do anymore.
• 0:48
  Speaker 1: It's about what it's actually doing in production and, frankly, how often it's breaking.
• 0:52
  Speaker 0: So today we've got a lot to get through.
• 0:54
  Speaker 0: We're talking agentic workflows, the hard reality of what some are calling vibe coding.
• 0:59
  Speaker 1: And some pretty tough truths about how product organizations are really structured versus how we like to pretend they are.
• 1:05
  Speaker 1: Let's
• 1:05
  Speaker 0: kick off with that big shift in AI.
• 1:07
  Speaker 0: For so long, AI was just this shiny feature.
• 1:10
  Speaker 0: You know, you'd have your app and then the ask AI button tucked away in the corner.
• 1:14
  Speaker 1: The bolt-on approach.
• 1:15
  Speaker 1: Yeah.
• 1:16
  Speaker 1: But based on the sources we're seeing, that era is just over.
• 1:21
  Speaker 1: Jeffrey Ollie makes this really compelling argument.
• 1:23
  Speaker 1: He says AI isn't a feature anymore.
• 1:26
  Speaker 1: It's becoming the infrastructure.
• 1:27
  Speaker 0: He has a specific phrase for that, right?
• 1:29
  Speaker 0: Moving from systems of record to systems of action.
• 1:33
  Speaker 0: Can we unpack that?
• 1:34
  Speaker 0: Because system record sounds like pretty much every piece of enterprise software out there.
• 1:38
  Speaker 1: Exactly.
• 1:39
  Speaker 1: Think of your CRM.
• 1:40
  Speaker 1: Its job is to just sit there and hold beta.
• 1:42
  Speaker 1: It's passive warehouse.
• 1:43
  Speaker 1: You put data in, you take it out.
• 1:45
  Speaker 1: Right.
• 1:45
  Speaker 1: Jeffrey Ollie's point is that AI is turning software into a system of action.
• 1:50
  Speaker 1: It doesn't just hold the data.
• 1:52
  Speaker 1: It reasons over it.
• 1:53
  Speaker 1: And it actually does something.
• 1:55
  Speaker 1: It executes the work.
• 1:56
  Speaker 0: And we are seeing some, I mean, staggering metrics on what that execution can look like.
• 2:01
  Speaker 0: Martin McDonough showed a case study that really stood out.
• 2:03
  Speaker 0: Oh,
• 2:03
  Speaker 1: the service design project one?
• 2:04
  Speaker 0: Yes.
• 2:05
  Speaker 0: They were analyzing this huge digital portfolio, which is normally just a complete slog.
• 2:10
  Speaker 1: It is.
• 2:11
  Speaker 1: It's weeks of reading documents, mapping user flows.
• 2:15
  Speaker 1: But Martin's team... used AI for that heavy lifting, the initial analysis.
• 2:19
  Speaker 0: And the results were pretty incredible.
• 2:22
  Speaker 1: They got the workload by eighty percent.
• 2:24
  Speaker 1: They went from twenty days of pure analysis down to just
• 2:27
  Speaker 0: four.
• 2:27
  Speaker 0: That's massive.
• 2:29
  Speaker 0: But here's my usual skepticism with stats like that.
• 2:32
  Speaker 0: Does that just mean the company gets to fire people or does the work actually improve?
• 2:35
  Speaker 0: Efficiency can be a loaded term.
• 2:37
  Speaker 1: That's the key question.
• 2:39
  Speaker 1: But in this case, it was explicitly about reallocating their time.
• 2:43
  Speaker 1: They took those sixteen days they saved and use them for synthesis, for actually talking to the client, understanding the nuance, and solving the real problem.
• 2:51
  Speaker 0: So less time reading, more time thinking.
• 2:53
  Speaker 1: Exactly.
• 2:54
  Speaker 0: Okay, so if we're handing off that reading and processing to an AI, we have to be able to trust it's doing a good job.
• 3:00
  Speaker 0: Right.
• 3:00
  Speaker 0: And this brings us to a major technical theme from the sources, AI evaluations, or ESOLs.
• 3:06
  Speaker 1: This is the piece that's missing for so many teams.
• 3:09
  Speaker 1: Akash Gupta has been very vocal about this.
• 3:12
  Speaker 1: He basically argues that product managers need to stop treating AI like a magic box.
• 3:17
  Speaker 0: And start treating it like an engineering problem.
• 3:19
  Speaker 1: Right.
• 3:19
  Speaker 1: You can't just prompt and pray.
• 3:21
  Speaker 0: Prompt and pray.
• 3:22
  Speaker 0: I love that.
• 3:23
  Speaker 0: It's definitely the default strategy for so many people.
• 3:25
  Speaker 0: You tweak the prompt, it looks good once, and you ship it.
• 3:28
  Speaker 0: And
• 3:28
  Speaker 1: then a user asks something slightly different, and the whole thing just falls apart.
• 3:32
  Speaker 0: So what does Akash say we should be doing instead?
• 3:35
  Speaker 1: He gets very tactical.
• 3:36
  Speaker 1: First, he says, to stop using these generic helpfulness scores, they're too subjective.
• 3:42
  Speaker 0: Was this helpful?
• 3:43
  Speaker 0: Yeah, that's a vanity metric.
• 3:45
  Speaker 0: A
• 3:45
  Speaker 1: complete vanity metric.
• 3:46
  Speaker 1: He advocates for what he calls binary judges.
• 3:49
  Speaker 1: Binary
• 3:50
  Speaker 0: judges.
• 3:50
  Speaker 0: Yeah.
• 3:51
  Speaker 1: Pass or fail.
• 3:51
  Speaker 1: Yeah.
• 3:52
  Speaker 1: Did the AI pull the correct document?
• 3:55
  Speaker 1: Yes or no?
• 3:56
  Speaker 1: Mm-hmm.
• 3:56
  Speaker 1: Did it hallucinate a number?
• 3:58
  Speaker 1: Yes or no?
• 3:59
  Speaker 1: Business logic is usually binary, so you need automated tests that check these constraints.
• 4:04
  Speaker 1: every single time you change a prompt.
• 4:06
  Speaker 0: It's basically unit testing for language models.
• 4:08
  Speaker 1: Precisely.
• 4:09
  Speaker 1: If you change the prompt to make the tone a bit friendlier, you need to know instantly if you just broke the core logic.
• 4:16
  Speaker 0: Okay, that makes sense.
• 4:17
  Speaker 1: But then he adds another layer of that.
• 4:20
  Speaker 1: Well, I think a lot of PMs are going to hate this part.
• 4:22
  Speaker 1: He insists on manual error analysis.
• 4:25
  Speaker 1: He says you have to sit down and read, one hundred traces.
• 4:28
  Speaker 0: A hundred conversations, line by line.
• 4:30
  Speaker 1: Yeah, a hundred actual AI conversations.
• 4:33
  Speaker 0: Reading through a hundred logs of JSON data, that sounds like absolute torture.
• 4:37
  Speaker 1: It does sound tedious, but his point is, you spot patterns that you would never, ever find on a dashboard.
• 4:43
  Speaker 1: You start to see why the model is getting confused.
• 4:45
  Speaker 1: You realize, oh, it fails every time a user asks about pricing in euros.
• 4:49
  Speaker 1: You
• 4:49
  Speaker 0: can't fix what you don't actually see.
• 4:51
  Speaker 1: You can't fix what you don't read.
• 4:53
  Speaker 0: Okay, so we've got the infrastructure, we've got the evaluation methods.
• 4:57
  Speaker 0: Now let's talk about what we're actually building.
• 4:59
  Speaker 0: The buzzword for twenty-twenty-six is, without a doubt, agents.
• 5:04
  Speaker 0: But Paul Wehran points out that most people are just building slightly better chatbots and calling them agents.
• 5:09
  Speaker 1: There's a huge difference.
• 5:10
  Speaker 1: Palos lays out a seven-step process for building real agents, and it goes way beyond just text in, text out.
• 5:18
  Speaker 1: Starts with a system.
• 5:18
  Speaker 1: prompt, sure, but then you have to pick a reasoning model.
• 5:21
  Speaker 1: He mentions the GPT-V family.
• 5:23
  Speaker 1: And this is the crucial part.
• 5:25
  Speaker 1: Give it access to tools.
• 5:27
  Speaker 0: Tools, as in it can actually check a calendar or query a database or maybe send an email.
• 5:33
  Speaker 1: Exactly.
• 5:33
  Speaker 1: An agent has memory.
• 5:35
  Speaker 1: It can orchestrate a plan.
• 5:37
  Speaker 1: It doesn't just answer a question.
• 5:38
  Speaker 1: It solves a multi-step
• 5:40
  Speaker 0: problem.
• 5:40
  Speaker 1: It has agency.
• 5:42
  Speaker 0: but this is where we hit that classic wall.
• 5:44
  Speaker 0: You can build the smartest agent in the world, but Raina Monforti brings up this very human problem.
• 5:50
  Speaker 0: She says the technology isn't the blocker.
• 5:52
  Speaker 1: It's trust.
• 5:52
  Speaker 0: It's trust.
• 5:53
  Speaker 0: Yeah,
• 5:53
  Speaker 1: this is a fascinating story.
• 5:55
  Speaker 1: She shared an example of an ops team.
• 5:57
  Speaker 1: Management rolls out this fancy new LLM tool to automate their workflow.
• 6:01
  Speaker 1: Technically, it's perfect, but adoption is totally stuck at twenty percent.
• 6:05
  Speaker 1: The team just wouldn't use it.
• 6:07
  Speaker 0: Were they worried it was going to hallucinate and make them look bad or take their jobs?
• 6:11
  Speaker 1: A bit of both, I think, but mostly just a fear of the black box.
• 6:14
  Speaker 1: They didn't understand how it worked.
• 6:16
  Speaker 1: So the fix wasn't technical at all.
• 6:18
  Speaker 1: They changed the workflow to include human in the loop steps.
• 6:22
  Speaker 0: So they put a person at a key approval point.
• 6:24
  Speaker 1: Yes.
• 6:25
  Speaker 1: They designed specific points where a human had to review and approve the AI's work.
• 6:30
  Speaker 0: And what happened to adoption?
• 6:32
  Speaker 1: Shot up to one hundred percent.
• 6:34
  Speaker 1: By making the automation less automated, they actually made it usable.
• 6:38
  Speaker 1: It gave the team psychological safety and they felt like they were still in control.
• 6:41
  Speaker 0: That connects perfectly with Jacqueline Reinhart's point on security.
• 6:45
  Speaker 0: Because if we're building these agents, that can take real action.
• 6:49
  Speaker 0: delete files, read emails.
• 6:52
  Speaker 0: Trust isn't just a feeling, it's an architecture.
• 6:55
  Speaker 1: You can't just hope the AI won't go rogue.
• 6:58
  Speaker 1: Jacqueline argues for a hybrid architecture.
• 7:00
  Speaker 1: Yeah.
• 7:01
  Speaker 1: Use the gen AI for the reasoning part, the creative thinking.
• 7:04
  Speaker 0: But you wrap it in what she called deterministic AI.
• 7:06
  Speaker 0: Right.
• 7:07
  Speaker 1: Deterministic meaning hard-coded rules, old-fashioned code.
• 7:10
  Speaker 1: You don't ask the LLM, hey, do you think this user should see this sensitive file?
• 7:14
  Speaker 0: No, you check the permissions database.
• 7:16
  Speaker 1: Exactly.
• 7:17
  Speaker 1: You put the AI in a straight jacket of strict rules.
• 7:20
  Speaker 1: That's how you prevent things like prompt injection.
• 7:22
  Speaker 1: You have to separate the brain from the keys to the castle.
• 7:25
  Speaker 0: That feels like a non-negotiable step if agents are going to do any real work.
• 7:29
  Speaker 0: Now, speaking of work, this whole shift is causing a bit of an identity crisis for product managers.
• 7:36
  Speaker 0: We've got this new term floating around, vibe coding.
• 7:39
  Speaker 1: I love this term, vibe coding.
• 7:41
  Speaker 1: It just, it captures the moment perfectly.
• 7:44
  Speaker 0: It sounds like something a DJ would do, not a developer, but it's a real thing, right?
• 7:48
  Speaker 1: Oh, it's very real.
• 7:50
  Speaker 1: Lenny Raczitzki put a spotlight on this with Xavier Arnovitz from Meta.
• 7:54
  Speaker 1: Now, Xavi is a non-technical PM.
• 7:56
  Speaker 1: He says he's scared to even look at a GitHub repo.
• 7:59
  Speaker 1: But with tools like Cursor and Claude Code, he's building real functional products.
• 8:04
  Speaker 1: So
• 8:04
  Speaker 0: the idea is you just tell the AI the vibe you want, make this pop, add a login screen, connect this to Stripe, and it just does it.
• 8:11
  Speaker 1: It's incredibly powerful for prototyping, but a very strong counterargument is emerging.
• 8:17
  Speaker 1: Aitir Abdul Raf warns that we're confusing building with engineering.
• 8:21
  Speaker 0: Okay, what's the distinction he's making?
• 8:23
  Speaker 0: If the app works, does it really matter how it was made?
• 8:33
  Speaker 1: But what happens when you need to integrate with some legacy API, or the system architecture needs to scale?
• 8:38
  Speaker 1: Yeah,
• 8:38
  Speaker 0: when the app breaks, you can't just tell the AI, the vibe is off, please fix it.
• 8:42
  Speaker 0: You need to know how to read an error log.
• 8:44
  Speaker 1: You need to know what a four-oh-four is.
• 8:46
  Speaker 1: ATR says PMs need to understand APIs and system design more than ever.
• 8:50
  Speaker 1: Otherwise, you're just creating a huge mess that a real engineer has to come in and clean up later.
• 8:55
  Speaker 1: David
• 8:55
  Speaker 0: Pereira had a great take on this.
• 8:56
  Speaker 0: He said, vibe coding doesn't fix bad product thinking.
• 9:00
  Speaker 1: That is the bottom line.
• 9:01
  Speaker 1: It doesn't matter if you use Python or some magic pattern tool for prototyping.
• 9:06
  Speaker 1: The code is just the delivery mechanism.
• 9:09
  Speaker 1: If you don't understand the user problem, you're just building the wrong thing faster than ever before.
• 9:14
  Speaker 0: You're just scaling bad decisions.
• 9:16
  Speaker 1: And Jayzer Bratzidev points out this very real danger for AIPMs.
• 9:21
  Speaker 1: They get so sucked into the building, the tweaking, the debugging.
• 9:24
  Speaker 1: But
• 9:24
  Speaker 0: they stop doing strategy.
• 9:26
  Speaker 1: They become the engineer they are trying to avoid being.
• 9:29
  Speaker 1: And strategy is always the first thing to go when you're busy firefighting.
• 9:33
  Speaker 0: Which is a perfect segue to our next big theme, the organization itself.
• 9:38
  Speaker 0: Because you can have the best PMs, the smartest AI, but if the company structure is broken, it just doesn't matter.
• 9:45
  Speaker 1: It really doesn't.
• 9:46
  Speaker 1: Yeah.
• 9:46
  Speaker 1: And we're seeing some interesting pushback against that ideal product model.
• 9:51
  Speaker 1: For years, the story was all about empowered teams.
• 9:54
  Speaker 0: Give the team a goal and get out of the way.
• 9:56
  Speaker 0: Right.
• 9:56
  Speaker 1: But Aunt Murphy points out.
• 9:58
  Speaker 1: that in twenty twenty five, and now in twenty twenty six, we've seen this huge swing back to what he calls founder mode.
• 10:04
  Speaker 0: Founder mode.
• 10:05
  Speaker 0: That sounds like micromanagement with better branding.
• 10:07
  Speaker 1: It kind of is.
• 10:08
  Speaker 1: It's executives taking top-down control.
• 10:10
  Speaker 1: They're basically saying, I don't care about your discovery process.
• 10:13
  Speaker 1: I want this feature shipped by Tuesday.
• 10:16
  Speaker 1: It's a reaction to feeling like they've lost control.
• 10:18
  Speaker 0: So the autonomy of the product team is actually shrinking in some places.
• 10:21
  Speaker 1: In many places, yes.
• 10:23
  Speaker 1: The focus shifts entirely to delivery metrics.
• 10:25
  Speaker 1: Output over outcome.
• 10:27
  Speaker 1: How many story points did we ship?
• 10:29
  Speaker 1: Not, did we actually move the needle for the business?
• 10:32
  Speaker 0: That sounds incredibly depressing for anyone who actually loves building products.
• 10:37
  Speaker 0: But is there a way to do this right?
• 10:39
  Speaker 0: How do you implement a good operating model without it just collapsing?
• 10:44
  Speaker 1: Darren Emory has some very practical advice here.
• 10:46
  Speaker 1: He says, the biggest mistake is the Big Bang transformation.
• 10:50
  Speaker 1: You know, the all hands email that says, starting Monday, we are all empowered squads.
• 10:54
  Speaker 0: And for the next six months, nothing works because no one knows who to talk to.
• 10:58
  Speaker 1: Exactly.
• 10:59
  Speaker 1: Deer and suggests a pilot.
• 11:01
  Speaker 1: Start small.
• 11:02
  Speaker 1: One value stream.
• 11:04
  Speaker 1: One outcome.
• 11:05
  Speaker 1: One empowered team.
• 11:06
  Speaker 1: And give them ninety days.
• 11:08
  Speaker 1: Prove that this way of working actually works.
• 11:11
  Speaker 1: Then you expand.
• 11:12
  Speaker 0: Prove it before you scale it.
• 11:13
  Speaker 0: Seems logical.
• 11:14
  Speaker 1: And Sven van Horbeek adds a crucial first step to that.
• 11:16
  Speaker 1: He says, don't even start with the teams.
• 11:18
  Speaker 1: Start with the WHY.
• 11:20
  Speaker 1: Why are we doing this?
• 11:21
  Speaker 1: Is it to innovate faster, reduce churn?
• 11:24
  Speaker 1: If you can't state the business reason, you're just rearranging deck chairs.
• 11:27
  Speaker 0: It's changed for the sake of a new org chart.
• 11:30
  Speaker 1: And Melissa Apple points out another classic failure mode, which is trying to change the product team in a vacuum.
• 11:36
  Speaker 0: Oh, this is so common.
• 11:37
  Speaker 0: You tell PMs, your mini CEOs know.
• 11:40
  Speaker 0: But engineering is still run like a ticket-taking service desk.
• 11:43
  Speaker 1: And sales is still out there selling features from a roadmap a year in advance.
• 11:47
  Speaker 0: And the PM is just crushed in the middle.
• 11:49
  Speaker 0: They get all the responsibility, but none of the actual authority.
• 11:53
  Speaker 1: Melissa's point is that it's an ecosystem.
• 11:55
  Speaker 1: If your partners in sales and engineering aren't aligned, the model will fail, period.
• 12:01
  Speaker 1: You can't be an agile island in a waterfall ocean.
• 12:03
  Speaker 0: Afonso Malofranco had a great way of putting this.
• 12:06
  Speaker 0: He warned against the romantic narratives of big tech.
• 12:10
  Speaker 1: Yes, we all look at Spotify or Google and see these perfect autonomous pods.
• 12:15
  Speaker 1: But Afonso reminds us that those pods are propped up by a massive support structure, program managers, ops teams, platform teams.
• 12:23
  Speaker 1: You can't just copy the squad model without the scaffolding.
• 12:26
  Speaker 0: So how does a company know if they're actually ready for this kind of shift?
• 12:30
  Speaker 1: Dave Baines offers a health check.
• 12:32
  Speaker 1: It's basically diagnostic to see if your day-to-day reality matches your ambition.
• 12:36
  Speaker 1: It forces some really honest self-reflection.
• 12:39
  Speaker 0: Okay, let's bring this down to the individual.
• 12:41
  Speaker 0: We've talked tech, we've talked org charts, but what about the PM at their desk on a Monday morning?
• 12:47
  Speaker 0: How do they survive all this?
• 12:49
  Speaker 1: First, they have to reclaim their time.
• 12:52
  Speaker 1: Stephanie Liu shared a really alarming statistic.
• 12:55
  Speaker 1: She says, seventy percent of a PM's week is reactive work.
• 12:59
  Speaker 0: answering slacks, putting out fire, sitting in meetings they don't need to be in.
• 13:03
  Speaker 1: Seventy percent.
• 13:04
  Speaker 1: That leaves almost no time for deep strategic thinking.
• 13:09
  Speaker 1: Stephanie argues that leadership has to actively protect that focus time.
• 13:13
  Speaker 1: Maybe two hours of uninterrupted time every other day.
• 13:16
  Speaker 0: That sounds like a dream, but in one of those founder mode companies, that's got to be a tough sell.
• 13:20
  Speaker 0: Why aren't you answering my messages?
• 13:21
  Speaker 1: It's hard, but the alternative is just being a busy fool.
• 13:25
  Speaker 1: Which brings us to prioritization.
• 13:27
  Speaker 1: Groshan Gupta has some great advice on handling stakeholders.
• 13:31
  Speaker 0: The classic struggle.
• 13:32
  Speaker 0: Sales wants a feature, marketing wants a launch date.
• 13:35
  Speaker 0: And the PM just says, I'll try.
• 13:36
  Speaker 1: And I'll try is the worst possible answer.
• 13:38
  Speaker 1: It just erodes your credibility.
• 13:40
  Speaker 1: Groshan says you have to treat feedback as decision input, not a command.
• 13:44
  Speaker 0: So how do you say no without, you know, getting fired?
• 13:48
  Speaker 1: You reframe the question.
• 13:50
  Speaker 1: When a stakeholder asks for feature X, You don't just say no.
• 13:54
  Speaker 1: You ask, OK, to make space for X, what are we going to drop?
• 13:57
  Speaker 1: Or you ask, which of our OKRs does this move forward?
• 14:01
  Speaker 0: You make them do the trade-off math with you.
• 14:04
  Speaker 1: Exactly.
• 14:05
  Speaker 1: You turn it from a simple request into a negotiation about value.
• 14:09
  Speaker 1: It forces them to see that your time is an infinite.
• 14:12
  Speaker 0: Marlon Davis points out that a lot of this confusion just comes from mixing up product management with project management.
• 14:18
  Speaker 1: A distinction that gets blurred all the time.
• 14:20
  Speaker 1: Marlon's clarification is simple.
• 14:22
  Speaker 1: Product owns the value in the direction.
• 14:24
  Speaker 1: What are we building and why?
• 14:26
  Speaker 1: Project owns the delivery.
• 14:27
  Speaker 1: When is it coming and how do we execute?
• 14:29
  Speaker 1: And
• 14:29
  Speaker 0: when the PM is forced to become a project manager,
• 14:32
  Speaker 1: you get products that ship perfectly on time under budget and then fail silently in the market because nobody actually wanted them.
• 14:38
  Speaker 1: We executed
• 14:39
  Speaker 0: the wrong plan perfectly.
• 14:40
  Speaker 0: You got
• 14:40
  Speaker 1: it.
• 14:41
  Speaker 0: Finally, Tim Herbig offers a piece of advice that I think applies to, well, everyone in this industry.
• 14:46
  Speaker 1: Drop the jargon.
• 14:47
  Speaker 1: Yes,
• 14:47
  Speaker 0: please.
• 14:48
  Speaker 1: We hide behind these complex terms.
• 14:50
  Speaker 1: Hypothesis validation.
• 14:52
  Speaker 1: Discovery sprints.
• 14:53
  Speaker 1: Tim says just use plain English instead of we need to run a validation experiment.
• 14:57
  Speaker 1: Just ask.
• 14:58
  Speaker 0: How can we find out if we should build this?
• 15:01
  Speaker 1: It's such a simple, powerful question.
• 15:03
  Speaker 1: It cuts through all the noise.
• 15:04
  Speaker 1: Yeah.
• 15:05
  Speaker 1: And really, I think that sums up the whole theme of these last two weeks.
• 15:08
  Speaker 0: How do you mean?
• 15:09
  Speaker 1: Well, think about what we've discussed.
• 15:11
  Speaker 1: AI evils that use binary yes or no judgments, product models that start with the why instead of the process, and language that drops the jargon.
• 15:22
  Speaker 1: It's all about a search for clarity.
• 15:24
  Speaker 1: We're moving away from the hype and trying to get to the hard reality of building things that are actually valuable.
• 15:29
  Speaker 0: That's a fantastic summary.
• 15:31
  Speaker 0: We've covered a lot of ground today from AI as systems of action to the dangers of vibe coding and the ongoing struggle for strategy in a founder mode world.
• 15:41
  Speaker 1: It's a complex landscape for sure, but I feel like the path forward is clearer than it's been in a while.
• 15:45
  Speaker 0: I couldn't agree more.
• 15:47
  Speaker 0: And for everyone listening, here's a final thought for you to chew on.
• 15:50
  Speaker 0: We talked about how AI is taking over the execution, the coding, the analysis, all the grunt work.
• 15:56
  Speaker 0: If the doing becomes automated, your value as a human is entirely in the deciding.
• 16:01
  Speaker 0: So the question is, are your decision making skills sharp enough to be your only marketable asset?
• 16:08
  Speaker 1: That is the uncomfortable question for twenty twenty six.
• 16:10
  Speaker 0: If you enjoyed this episode, new episodes drop every two weeks.
• 16:14
  Speaker 0: Also check out our other editions on ICT and tech, artificial intelligence, cloud, sustainability and green ICT, defense tech and health
• 16:21
  Speaker 1: tech.
• 16:22
  Speaker 1: Thanks for listening and don't forget to subscribe so you don't miss our next deep dive.