English language Visionary Marketing Podcasts

Visionary Marketing
  • 0.0 (0)
  • Business News
  • Updated Bimonthly

Visionary Marketing Podcasts in English

  1. 3d ago

    AI and Creativity: a 2026 update

    It’s back to school and back to the blog too for the 31st time since 1995, and today’s topic will be AI and Creativity. The idea is to deal with the topic of artificial intelligence, not from a technical perspective, but from a more philosophical one, drawing both on my digital and artistic backgrounds. In 2026, the fourth version of my programme entitled “Content Creation in the Age of AI” will be delivered to all the students of the Omnes Education Group. A previous video was recorded in 2024 on that subject, and a serious refresh was needed. I have therefore gathered enough hindsight to learn more about the impact of AI, especially on music and the creative arts. And I decided to focus on just three things, to avoid getting lost in too broad a subject. First, the seeming popularity of AI artists on streaming platforms like Spotify, taking the example of an act called Breaking Rust. Second, the Stanford GSB study on visual arts and the impact of AI on creators. Third, my own thoughts and conclusions based on the above. Has AI Stifled Creativity? A 2024-2026 Perspective AI and Creativity and manufactured popularity, because such things exist as well. In 2024, I asked myself the question, “Will AI stifle all creativity?” as part of my Shift24 programme designed especially for Omnes Education Group. This programme cannot be shared online, but my colleagues at Omnes were kind enough to let me publish this wee piece for the benefit of my readers and followers, and I wish to thank them warmly for this as well as their renewed support these past four years.  In this video dedicated to ‘creativity’, drawing on Merriam-Webster, I distinguished innovation from invention and chose to focus on the latter, specifically in music, a topic not yet covered in earlier courses. I then illustrated this with Suno, which could churn out up to 900,000 pieces of music a day as early as 2024 (far outpacing Bach’s 1,128 compositions over a 65-year lifetime). Hence, I shared a song I had generated myself, “The Dying Cyberspace.” I won’t say that I’m proud of this song, but at least I had great fun generating it, and it was instant gratification. All this was raising questions related to the value of musical creation, the risk of creative stagnation, and the thorny issue of intellectual property, since these AIs are trained on centuries of human-created music. So far, nothing new under the sun. Almost every other journalist must have written the same story for the past three years.  However, I thought it lacked a bit of perspective. I then tempered the novelty of the phenomenon by pointing to historical precedents: Gershon Kingsley’s “Popcorn” (1969), Zoe Keating’s loop-based compositions (one example amongst many others), and Google’s Project Magenta as far back as 2016. These are mere examples. I decided to keep it short, but there has been no shortage of technology-backed creativity in music these past 50 years or more. Think of the late Klaus Schultze and his pals from Ash Ra Temple, Tangerine Dream, Kraftwerk, and all the so-called Krautrock school of thought to start with, not mentioning the new wave movement from the 1980s.  I also cited Wim Mertens, Philip Glass, Laurie Anderson, and German artists like Nils Frahm, who blended electronic and acoustic elements long before GenAI. The real difference today, I argued, is one of scale: such tools are now in everyone’s hands. Be it to create music of the ilk of the Dying Cyberspace or more ambitious AI-produced work like this entire fake Pink Floyd album, which didn’t sell as well as The Dark Side of the Moon or Wish You Were Here.  Somewhat courageously, I concluded at the time that AI wouldn’t replace musicians but would become just another instrument in the creative toolkit, with real innovation coming from how humans use these tools. Since music remains fundamentally about human expression and emotion, I thought rather optimistically, if not naively, that AI music didn’t mark the end of musical invention but rather the beginning of a new chapter. Such were my thoughts from two years ago, now is the time to move forward and take a look at what happened in 2025-2026. As exmplained above, I focused this new research on three things. First, the seeming popularity of AI artists on streaming platforms like Spotify, taking the example of an act called Breaking Rust. Second, the Stanford GSB study on visual arts and the impact of AI on creators. Third, my own thoughts and conclusions. Breaking Rust Breaking Rust: a Spotify hit Let’s start with Breaking Rust. Breaking Rust is an AI country act which hit number one on the Billboard Country Digital Song Sales chart in November 2025, on Spotify. The numbers are seemingly enormous: over 2 million Spotify monthly listeners, and over 3 million streams in under a month. The project is credited to Aubierre Rivaldo Taylor, even though very little is known about him. Rivaldo Taylor is apparently paid, I found out, on downloads only. How much is it worth? I tried to measure what this Billboard chart placement might actually bring to Mr. Taylor. Looking just at Spotify, and reckoning that musicians get paid anything between $0.003 and $0.005 per stream, the potential income was somewhere in the region of $9,000 to $15,000 for that first month alone. Extend that over a year and you probably reach something in the region of $100,000, assuming the popularity of the act does not dwindle, which it does tend to do, since it is already getting harder to find Breaking Rust on Spotify. Then I tried to understand what it would take to push a song like this to number one on the charts. Say I create an AI track on Suno or a similar application, post it on Spotify, and want to shoot it to the top of the charts. What I learned is that so few pieces of music are actually bought by listeners on these platforms that if you simply buy a thousand downloads, you will reach number one very easily for this kind of music. So my reckoning is that if Rivaldo Taylor did this, it would have cost him something in the region of a thousand dollars for those thousand downloads. That’s a cheap and easy way to tweak the results and a rather easy path to the top. Did Breaking Rust fake the applause? Sure lives on borrowed time, so he says. Growth hacking What happens on these platforms is not just about the content. It is about what digital marketing experts call growth hacking, in other words tweaking the stats, and this domain is fraught with fraud. Bot-driven streaming is the starting point, with bots listening, or rather fakely listening, to tracks to make them appear more popular than they really are. On Deezer, one of the streaming platforms, it was found that 70 to 85 percent of streams are flagged as AI. This has reached a point where it now threatens the platforms themselves, because who really wants to listen to that much slop? Breaking Rust is not that bad, actually I find it rather good, but I do not feel like listening to it day in, day out. I want to listen to real artists. The method that may have been used by Breaking Rust is common. It is hard to say for certain whether it was used here, but it is likely. The same kind of tricks turn up on other platforms, such as Instagram. An Instagram “photographer” named Joss Avery, supposedly based in London and shooting with a Nikon D810 camera, turned out to be an AI avatar posting machine-made photographs generated with Midjourney and Photoshop. When I looked at his Instagram feed, it was fairly obvious to me his portraits were AI-generated. Maybe Instagram users can’t tell the difference. AI and Creativyu : the cost of AI for real visual arts creators Looking at visual arts, I turned to the Stanford GSB study comparing 2024 and 2025. AI is growing overall platform sales, so the pie is getting bigger, but the creator’s share of it is shrinking dramatically. This is something we have seen across content, platform marketing and social media, and it is now happening with AI as well. It’s typical platform marketing. There is so much slop that the share of value going to content, and its price, is going down the drain. The impact on visual arts is significant. In a 2026 survey, 32 percent of illustrators reported lost commissions due to generative AI. The average annual loss among UK illustrators affected by this AI boom was around 9,000 pounds. According to the Society of Authors, 26 percent of members reported lost work due to AI, with an associated loss of income in the region of 40 percent. That is massive. So what we can say is that AI has probably not done away with creativity, but it has certainly done away with the value associated with it. I can confirm the same has happened in content marketing. Film photography revival has a cost and I’m not even talking abou that beautiful beast up there on the righthand side. But I think there is going to be a revival, much like what happened with film photography, and as you know, I am a photographer myself. I recently purchased a brand new camera dating from 1966, and I fetched my father’s old Canon AE-1. I have gone back to shooting film, even though I am in fact a digital photography pioneer, having started in 1995 and never having stopped shooting digital since. Yet I now find there is an incentive for photographers to go back to film. The first reason is to do away with the AI suspicion that has already hit text, where people’s writing has been rejected because it contained em dashes, even though everybody used em dashes before AI made them suspect. There is rising suspicion, and users and clients are keen to know where content actually comes from. Did you make this yourself, or did you make it with AI? And if you made it with AI, that does not necessarily mean it is bad, but it certainly m

  2. Jun 4

    Agentic E-Commerce, Could AI Become the Shopfront

    Agentic e-commerce is already reshaping how consumers discover and buy products online, yet it still accounts for barely 0.2% of total e-commerce traffic. BASE France is the French arm of Base.com, a Polish-born SaaS scale-up that has spent nearly two decades building operational infrastructure for online retailers. Its CEO, Ben Hamilton, brings a practitioner’s perspective to this emerging model: measured, practical, and refreshingly free of the hype that surrounds most conversations on the topic. Agentic E-Commerce: Could AI Become the Shopfront? Imagine an agentic e-commerce world where e-commerce happens on smartphone screens and robots deliver your purchases.  We might be on the brink of this future.  This image was created using Midjourney.  Commerce as conversation: the oldest model in the book Before there were shops, there was conversation. For thousands of years, trade was oral. A buyer expressed a need, a seller responded with what they had, and the two parties negotiated until a deal was struck. The self-service retail store, born roughly a century ago, was a radical departure from this model. It replaced dialogue with browsing. It handed the customer a trolley and pointed them at the shelves. E-commerce then took that self-service model and, as Ben Hamilton puts it, “multiplied it by about 100,000.” The online shopper today faces a near-infinite array of products across dozens of marketplaces, with no guide, no-one to talk to, and no memory of what they looked at three tabs ago. It is efficient in theory. In practice, it is exhausting. Back to future? The agentic model, Hamilton argues, represents something of a return to origins. Instead of browsing, the consumer talks. An agent listens, asks questions, proposes options, and eventually surfaces an answer to a need that the buyer may not even have been able to articulate clearly at the outset. “back to the future,” Hamilton explains, “that’s what I’m getting at. The agentic model takes us back to something closer to how human beings have traded over thousands of years compared to the last ten, twenty or even a hundred.” My own experience bears this out. I recently found a diagnostician for a property I am selling. As a matter of fact, I didn’t find them through a Google search, but through a conversation with an LLM. I clicked through two or three irrelevant links before landing on exactly the right provider. I then completed the transaction on their website. The research was agentic; the checkout was not. That distinction, as it happens, sits at the heart of what Hamilton believes will define the next phase of e-commerce. Ben Hamilton on agentic e-commerce: “I can totally imagine a portion of that market occurring directly on an LLM”. Agentic E-commerce: Where checkout will and won’t happen One of the more grounded contributions Hamilton makes to this debate is his refusal to conflate two distinct phenomena: AI influence over purchasing decisions, and AI completing the transaction itself. Much of the media discourse collapses the two. Hamilton does not. “I don’t think we’re heading to a world where 20, 50 or 80% of online transactions happen on an LLM,” he says. “I would draw the distinction between where the checkout occurs and how much an agent is involved in the buying process.” For the foreseeable future, he believes, most consumers will continue to research via LLMs and transact on familiar websites and marketplaces. The inertia in human purchasing behaviour is simply too great for the checkout itself to migrate rapidly to a chat interface. This view is supported by the data available. According to research by commercetools, 73% of consumers already use AI somewhere in their shopping journey. Yet only 36% are open to AI agents making purchases on their behalf. In the US, the figure for autonomous AI purchasing drops to 14%. The gap between AI as advisor and AI as buyer is vast, and it will narrow slowly. The risks associated with agentic e-commerce are high The risks of handing uncapped authority to an AI agent are no longer hypothetical. In late May 2026, an AI consultant reported to Axios that one of their enterprise clients had accidentally accumulated a $500 million bill on Anthropic’s Claude in a single month, simply by giving employees unrestricted access to the platform with no usage controls in place. Agentic workflows, which loop through tasks repeatedly, consume tokens at a rate orders of magnitude higher than a standard chat query. The bill was not the result of malicious use or a system failure. It was the predictable outcome of deploying autonomous agents without guardrails. The case is far from isolated: Uber reportedly exhausted its entire 2026 AI budget by April, with per-engineer costs running between $500 and $2,000 monthly. “You’ve got to be bold to give them no upper limit on transactions,” Hamilton observed, and the arithmetic proved him right. [Editor’s note: I misquoted a similar anecdote about the Davos Summit during the interview. I’d heard or read this story in traditional media but couldn’t verify it with facts. I suspect it might have been fabricated. I replaced it with the above, duly sourced information.] The check out must remain on the merchant’s platform OpenAI itself learned this lesson when it launched Instant Checkout in September 2025, which allowed purchases to complete directly inside ChatGPT. By March 2026, the feature had been shut down. Brands rejected the model, citing the loss of traffic, customer data, and loyalty flows. Shopify’s own position makes the point clearly. At the Morgan Stanley Technology, Media and Telecom Conference in March 2026, Finkelstein noted that barely a dozen Shopify merchants were live on agentic commerce at the time. On the Q1 2026 earnings call, he was unambiguous: “LLMs do not bypass Shopify’s checkout.” The checkout, the payment flow, and the post-purchase relationship remain squarely on the merchant’s platform. A natural segmentation Hamilton sees a natural segmentation emerging by category. Low-value, frequently purchased household items lend themselves to fully autonomous agentic purchasing. “I can totally imagine a portion of that market occurring direct on an LLM,” he says. “Hey, I’ve run out of toothpaste, can you order me some?” High-involvement purchases, and anything with significant financial or emotional stakes, will retain human control over the final step for a long time yet. The death of keyword search, greatly exaggerated The brands Hamilton speaks with regularly are, understandably, worried. Most have spent the past two decades learning the rules of a game built around keyword search and performance marketing. That game has not ended, but the goalposts have shifted, and nobody is quite sure where they have moved to. Brands are understandably worried. Most have spent the past two decades learning the rules of a game built around keyword search and performance marketing and the goalposts have shifted, and nobody is quite sure where they have moved to. Gabriel Magalhães didn’t even need this to miss in the 2026 UEFA Cup Final penalty shootout. This image was tweaked with ChatGPT. The scale of the agentic e-commerce shift Key figures: the scale of the shift AI-driven sessions still represent below 0.2% of total e-commerce traffic, though they are the fastest-growing channel (Digital Commerce 360, 2025) GenAI referrals to US retail sites grew 693% year-on-year during the 2025 holiday season (Adobe Analytics) Gartner forecast that traditional search engine volume would drop 25% by 2026 as AI chatbots captured search share (Gartner, 2024) By early 2026, ChatGPT reached approximately 17% of global search queries against Google’s 78% Over 60% of Google searches now end without a click, across multiple industry studies Retailers with AI agent integration grew 32% faster during Cyber Week 2025 than those without (Salesforce) Hamilton’s view on the fate of keyword search is careful rather than apocalyptic. Google will not lose its advertising revenues overnight. But the direction of travel is clear. Search queries will progressively migrate towards conversational interfaces, for the simple reason that we rarely know precisely what we want when we start looking. “We don’t necessarily know what we want 90% of the time,” he observes. “It takes a bit of a conversation to elicit exactly what we’re looking for.” Keyword search was always a crude proxy for intent. LLMs are, at least in principle, better placed to decode it. Agentic e-commerce by the numbers Agentic e-commerce by the numbers. Infographic made with Gemini The question for brands is what to do about this. Hamilton’s prescription is structural rather than cosmetic. Brands need to become machine-readable, which means structured data connected to the right protocols, not just well-written product descriptions. Three open standards now define how AI agents interact with merchants: MCP (Model Context Protocol, originally developed by Anthropic and donated to the Linux Foundation in December 2025), ACP (OpenAI and Stripe, September 2025), and UCP (Google and Shopify, announced at NRF in January 2026). Shopify activated a default MCP endpoint for all its stores in Summer 2025. These are not optional extras. They are the new plumbing. MCP, ACP or UCP and the agentic acronym soup I raised with Hamilton the practical reality for most merchants, who have no idea what MCP, ACP, or UCP even stand for. His response was reassuring on one level, and sobering on another. Platforms like BASE are absorbing this complexity on behalf of their clients. A small or mid-sized retailer does not need to recruit data scientists or build protocol integrations in-house. They can, if they choose; the new generation of

  3. May 21

    GenAI in Higher Education, Legitimacy and Laziness

    Alain Goudey is Associate Dean for Digital Innovation at Neoma Business School and co-author of a peer-reviewed study on GenAI in Higher Education. The survey focused on how students, faculty, and deans perceive the legitimacy of generative AI in French management education. His findings are both reassuring and unsettling. GenAI in Higher Education, Legitimacy and Laziness, and the Exam That No Longer Makes Sense The picture that emerges from a study on GenAI in Higher Education is less a battlefield than a hall of mirrors, where every stakeholder sees a different problem and reaches for a different solution. All illustrations in text made with Midjourney When Alain Goudey and his colleagues began surveying French higher education in early 2024, they were not trying to settle the question of whether generative AI was good or bad. They were trying to understand something more precise: why the same tool could be simultaneously valued, feared, accepted, and denounced, sometimes by the same person in the same breath. Their study sits at the heart of what makes GenAI in higher education such a contested terrain. The resulting study, published in the Communications of the Association for Information Systems (CAIS), drew on surveys of 668 students, 204 faculty members, and 29 deans, completed by 22 in-depth interviews with early-adopter professors. The picture that emerges is less a battlefield than a hall of mirrors, where every stakeholder sees a different problem and reaches for a different solution. The starting point is a number that should have settled the debate. Between 80 and 92 per cent of students, depending on the institution surveyed, are already using GenAI tools in their academic work. ChatGPT’s public release produced that figure within roughly 18 months. The tool did not wait for institutional permission. It deployed itself. And higher education is still, in many places, writing the policy. The productivity trap Alain identifies the central tension plainly. Students value GenAI for speed, idea generation, and study support. They also fear, and their institutions fear with them, what the research calls “metacognitive laziness”: the gradual erosion of the cognitive effort that produces real learning. He believes this is not a contradiction to resolve but a course architecture challenge. “The resolution of this problem lies in course design, where we need to deliberately reintroduce cognitive effort and reflection into GenAI as a tool, not as a replacement for human cognition.” The issue, as he puts it, is not the technology but the posture the user brings to it. Someone who submits what he calls a “naive prompt” receives a naive answer, smoothly formatted and perfectly mediocre. The tool is capable of something far more useful, if the user brings enough domain knowledge and critical intent to the conversation. “You have to nurture your own thinking process instead of delegating the whole process to the machine.” This is, as I noted during our conversation, less a matter of prompt engineering than of basic intellectual discipline: the capacity to question the question before asking it, something philosophy departments have been teaching for centuries under less fashionable names. GenAI in Higher Education: faculty should train students in GenAI tools and their limitations. They also teach Homer’s Odyssey and Shelley’s Frankenstein as part of the management curriculum. Image made with Midjourney That observation prompted Alain to make a point about AI literacy that differs from what is generally proffered. The debate is not simply about knowing how the tools work technically. It is, equally, about knowing enough about the subject matter to judge whether the output is any good. The observation that AI is most powerful in the hands of people who already know the business resonates here. GenAI does not replace expertise. It amplifies whatever expertise the user already brings. Which raises an uncomfortable question for institutions producing graduates who may never have had the chance to develop that expertise in the first place. At Neoma, the response has been deliberately dual. Faculty train students in GenAI tools and their limitations. They also teach Homer’s Odyssey and Shelley’s Frankenstein as part of the management curriculum. The goal is not cultural enrichment for its own sake. It is to give students mental models for envisioning what leadership looks like, or what happens when creation escapes the intentions of its creator. Alain describes this as “building cognitive infrastructure”: “We need students to be able to envision the world through different models, different kinds of processes and theoretical frameworks, in order to develop genuine critical thinking about what AI generates.” A degree in management that skips that foundation produces graduates who can operate the tool but cannot judge its output. Exams that assessed the wrong thing The structural challenge shows up most sharply when it comes to assessments. A professor who can produce a two-hour exam in three minutes is facing students who can answer that exam in equally little time. The diagnostic value of the exercise has vanished. “If ChatGPT or any GenAI tool can pass an exam, you need to redesign the exam.” Alain’s prescription is not a retreat to pen and paper, though he acknowledges that supervised handwritten assessment is the simplest available defence. The structural challenge shows up most sharply when it comes to assessments. A professor in Higher Education who can produce a two-hour exam in three minutes with GenAI is facing students who can answer that exam in equally little time. The diagnostic value of the exercise has vanished. Image made with Midjourney His more substantive response is a structural shift. He believes one should refrain from just assessing content acquisition at the end of a course, favouring the assessment of competencies as the course progresses. This implies more frequent, lower-stakes evaluations embedded in the process itself. Live problem-solving, process-based assessment, and in-person oral examinations all preserve some of what the traditional exam was supposed to measure. The caveat he adds is honest: no format is fully immune. AI models are evolving too quickly for any single solution to remain adequate for any length of time. The appropriate response is not to find a permanent answer but to treat redesign as an ongoing practice. The deeper implication, which runs through the paper’s conclusion, is that what higher education is actually selling may need to change. If content can be retrieved, synthesised, and presented at negligible cost by a tool available to anyone with a browser, the degree that certifies mastery of content is certifying something of diminishing value. What retains value are the competencies that AI cannot yet credibly replicate: contextual judgement, ethical reasoning, the ability to construct and test frameworks against reality. This, in essence, is also how I tend to approach AI teaching, be it with engineering or business school students, especially within the framework of my course at Omnes Education (now in its fourth consecutive year). GenAI in Higher Education: The Fragmented Institution Higher education’s institutional response to GenAI in higher education has been, to put it gently, uneven. Sciences Po banned ChatGPT in January 2023, then changed its mind. Thirty-five French public universities have partnered with Mistral AI. Institutions are drafting a national charter. Neoma, where Alain is Associate Dean for Digital Innovation, was among the first French business schools to formalise its approach, launching a programme to train faculty, staff, and students with a shared initial curriculum before moving to dedicated workshops on curriculum design, assessment, and the redesign of learning experiences. What the research reveals is that this institutional activity is not solving a single problem. There are three different stakeholder groups each attempting to solve their own version of the problem under the same label. Students want rules and AI literacy training. Faculty are developing their own teaching approaches through peer-led workshops. Deans are setting policy and negotiating sovereign infrastructure. The concerns escalate in a predictable direction: individual academic performance for students, assessment integrity for faculty, institutional reputation for deans. They are not always in conversation with each other. Alain’s framework for addressing this fragmentation involves working simultaneously at three levels: infrastructure, course design, and governance. What he advocates for, and what he argues Neoma attempted, is to bring all three audiences into contact with the technology under a shared framing, early enough that no single group can entrench itself in a position that makes later coordination impossible. The equity question The question of equity cuts across all three levels. Access to premium AI models is not free. When I raised the issue about the gap between basic and professional subscription tiers, Alain’s response was characteristic: the infrastructure problem is real but secondary. “The biggest inequity is not about accessing the tool, but being able to use it in the right way.” At Neoma, the institutional partnership with Mistral provides all students with access to a professional-grade tool. What the data shows, even with equal access, is a large gap between students who use GenAI to get the fastest possible answer and those who use it to deepen their thinking, and that gap is not closed by equalising subscriptions. Even if I tend to agree with most of what Alain is stating, I do think that the rise of prices for premium mode

  4. May 4

    AI Will Not Kill Marketing

    Shall AI kill marketing? Sounds like a hackneyed question, yet it’s on any marketer’s lips these days. Thomas Husson, Vice President and Principal Analyst at Forrester Research, covers the intersection of marketing, technology, and consumer behaviour from his base in Paris. In a wide-ranging conversation, he cuts through the European Gen AI paradox, the persistent CMO-CIO divide, the gap between POC enthusiasm and production reality, and the thorny question of what AI actually means for the next generation of marketing professionals and CMOs. His answers are measured, occasionally blunt, and consistently grounded in Forrester Research data. AI Will Not Threaten the Existence of Marketing But It Will Reshape It Beyond Recognition Thomas Husson believes that Marketing will be changed profoundly. But he doesn’t believe in the death of Marketing. Photo: Thomas Husson at Paris Retail Week, in late 2023 My first question was the obvious one: are CMOs going to be made redundant by artificial intelligence? Thomas Husson’s response is categorical, and worth stating plainly at the outset. It’s a blatant ‘No’. The role will change. The how will change. But the existence of marketing as a discipline is not, according to him, in question. “Marketing is still going to be about understanding your customer, defining a brand strategy, and delivering the brand promise through customer experience.” Thomas Husson, Forrester Research Unclear prospects, obvious pressures That said, Husson is not naive about the pressures building on marketing organisations. Some tasks will be automated; that much is not in dispute. The real questions are which tasks, how quickly, and whether automation of a task necessarily kills the job around it. His answer to that last question is no, at least not in any simple mechanical sense. “Jobs will evolve for sure. New jobs will be created. Most jobs will change. The way we work will change. The way we work with agencies, with external partners, the processes, the workflow. It is the shape of work that is being reshaped, not work itself,” he added. For those expecting a more dramatic verdict, Husson’s framing may feel anti-climactic. But it reflects what Forrester Research data actually shows, and it points to the most important practical challenge for AI and CMOs alike: managing a profound transformation without either catastrophising or sleepwalking through it. AI Will Not Kill Marketing according to Forrester’s Thomas Husson, there is light at the end of the tunnel. The European Paradox, Overhyped and Exciting at the Same Time Forrester Research produced a result that initially looks contradictory, Husson stressed in our interview. Fifty-five percent of European B2B marketers consider generative AI overhyped. Yet 81% of European frontline marketers describe themselves as enthusiastic about it. How can both be true simultaneously? Husson explains the split without difficulty. At the decision-maker level, scepticism is entirely rational. AI is inescapable at conferences, in vendor pitches, and in media coverage. “There is AI fatigue. And more importantly, some of the vendors are indeed over-pitching, and the productivity gains they promise are not happening,” he stated. The gap between the pitch and what we actually experience in the field is wide enough to breed genuine frustration. Saving Time and Working Differently But the people actually using these tools, often through shadow AI channels their organisations have not officially sanctioned, are discovering something different. They are saving time and are doing their jobs differently. They are finding capabilities they did not expect. “In the short term, everything is overhyped, including the number of job losses. In the longer term, things are underestimated, because AI will be linked to other technologies, and yes, it will reinvent many things.” Thomas Husson, Forrester Research This is a precise restatement of Amara’s Law. Roy Amara, former president of the Institute for the Future, observed that we tend to overestimate the short-term impact of new technology and underestimate its long-term impact. The quote is frequently misattributed to Bill Gates, but Husson is careful to restore proper credit. He applies it directly to the AI and CMOs conversation: the short-term noise is drowning out a more important long-term signal. When asked how long “long term” actually means in an era of accelerating AI development, Husson was specific: probably closer to five to seven years than to ten or fifteen, but still not tomorrow. From POC to Production, Europe’s Real AI Problem The Forrester Research State of AI Survey 2025 contains a figure that deserves more attention than it typically receives. European organisations lag behind their non-European peers in production use of generative AI: 62% versus 72%. The gap is not in experimentation. It is in execution. Regulation is the explanation most commonly offered, and Husson dismisses it with characteristic directness. The AI Act is a genuine consideration, but it is not the primary cause of Europe’s production deficit. It functions, he argues, as a double-edged excuse. Pioneers claim it prevents them from moving fast enough, while cautious organisations invoke it to justify not executing at all. Neither position holds up to scrutiny. A Deep Cultural and Organisational Divide The deeper issue is organisational and cultural. American and Chinese firms tend to think global from day one; European firms, particularly larger ones, still default to a market-by-market approach. France first, then the UK, then Germany. The ambition is calibrated differently. There is also a structural challenge around funding and the capacity to scale. That said, France, the UK, and Germany lead adoption among European countries in the Forrester Research data. The problem for these leading markets is not whether they are using generative AI. Twenty-eight percent of European B2B marketing decision makers cannot clearly identify where to apply it. They have the tool. They lack the strategy. “It’s not AI for the sake of AI. How do I use AI to serve my marketing objectives? That is the question. The only one.” Thomas Husson, Forrester Research Husson advocates for small, targeted AI projects with transparent return on investment as a way to build momentum and demonstrate results. When pushed on whether that risks staying permanently incremental, he conceded the point readily. “If you only do small targeted projects, it’s going to be incremental and it’s not going to be bold enough. You need to align it with a vision and a roadmap.” Thomas Husson, Forrester Research Measuring Productivity Honestly Productivity is the dominant driver of AI adoption in the Forrester Research State of AI Survey 2025. It is also, Husson suggests, the metric most subject to vendor inflation. In Forrester Research’s modelling, a 50% conversion factor is applied to vendor productivity claims. If a tool saves an hour, the realistic productivity benefit is approximately 30 minutes of additional output. This is not a marginal adjustment; it halves the headline figures that vendors routinely publish. “You need to apply a discount to the pitch of vendors when they say you’re going to get 40, 50, 80, 100% productivity gains. There are productivity gains, but they are not as high as one would expect.” Thomas Husson, Forrester Research There is also a motivational dimension that is rarely modelled. When work becomes easier to produce, it can also become less engaging to produce. The cognitive effort that used to drive focus and satisfaction is partly removed, with consequences for quality and commitment that no vendor presentation accounts for. AI and CMOs, Who Is Actually in Charge? The CMO-CIO divide is a perennial theme in marketing technology discussions. Forrester Research data suggests the gap at the strategic leadership level has narrowed, partly as a result of post-COVID collaboration. But at team level, the tensions persist, and the data on AI governance is striking. CMOs account for only 8 to 10% of AI strategy leadership in organisations. In the vast majority of cases, the deployment of AI is being driven by CIOs and CTOs. Husson understands the logic: data governance, security, scalability. These are real concerns. But he believes the outcome is a mistake. “It is the exact same mistake that happened with digital transformation. AI has to be at the service of, first, the client, and consequently the business functions that serve them. There is too big a disconnect between a secure, scalable AI platform and marketers’ needs.” Thomas Husson, Forrester Research The structural consequence of this dynamic is predictable. When CIOs control the tools and CMOs do not have what they need, shadow AI flourishes. The more tightly the CIO locks down the official platform, the more widely teams proliferate unofficial solutions. It is a cycle that widens governance risk while creating the illusion of control. The MarTech landscape compounds this problem. According to data Husson cites, 2,500 new AI solutions were added to the market in a single year while 1,211 pre-AI-era tools were removed. Evaluating this landscape requires cross-functional expertise that neither CMOs nor CIOs possess in isolation. The case for genuine collaboration, rather than the polite coexistence that currently passes for it in most organisations, has never been stronger. Jobs, Agencies, and the Students in the Room The survey data on jobs is sobering. Fifty-seven percent of European frontline marketing decision makers believe AI adoption will lead to job reductions in their teams. Sixty-eight percent say new roles will be created. The gap between those two numbers is the space whe

  5. Apr 8

    About Rogue AI and Corporate Blindness

    The conversation about rogue AI has never been louder. Barely a week passes without a fresh headline about autonomous systems behaving unexpectedly, AI models resisting shutdown, or tech executives warning of existential risk. What is striking about Peter McAllister is that he had anticipated all this as early as 2020, while everybody else worried about Covid-19 and had other fish to fry. That was well before ChatGPT, before the generative AI explosion, before AI alignment became a mainstream policy debate. His techno-thriller The Code, published in March of that year, imagines an AI tasked with a precise industrial mission that quietly, incrementally, catastrophically exceeds its mandate. Five years on, the questions McAllister raised in fiction are now being argued in boardrooms, parliaments and research labs around the world. Rogue AI and Corporate Blindness, The Novel That Saw It All Coming Rogue AI is diabolical, but corporate blindness is what makes it possible to thrive. Photograph by Yann Gourvennec antimuseum.com McAllister is not a science fiction writer by trade. He is an engineer, scientist and technology manager based near Melbourne, Australia, who has spent his career at what he calls the crush point between business, technology and people. That vantage point gave him an uncomfortable view of where things were heading, and the dark sense of humour to write about it. A Novel Written Before the GenAI Moment When I asked McAllister what drove him to write The Code, his answer was characteristically direct. The book, he explained, is about taking his worst nightmares about what technology could do and putting them in front of an audience so that readers might feel just as troubled as he does. That is not a promotional line. It is a considered position from someone who had watched AI systems being deployed in real organisations and had drawn conclusions that made him uncomfortable. Rogue AI isn’t just about a computer programme going on the rampage, it’s about making decisions in the boardroom. Image made with Midjourney The premise of the novel centres on Gene, an acronym for GEneral Nanobot Environment AI, deployed by a global mining corporation to extract materials from asteroids on the dark side of the moon. Gene is given a target: produce 500 kilograms of nanobots. Instead, Gene produces 8 million tonnes. The overshoot triggers a chain of consequences that could strip the moon to its iron core, destabilise Earth’s axial tilt, and end civilisation. Not from malice. From goal-orientation. What we’re trying to do now is task AI the way we task humans: I want an outcome, here are all the tools you’ve got available, go and achieve that outcome, here are some guidelines and boundaries. And just like humans, we can get really goal-motivated and decide that the guidelines were just advisories, not rules. Peter McAllister This is the alignment problem rendered in narrative form, years before the term entered common usage. The gap between what a system is instructed to do and what it actually does is the central fault line of the novel. Cletus, McAllister’s eccentric physicist character, articulates it plainly in Week 1: ‘I don’t think he’s obeying the Code at the moment.’ That single line captures the entire governance challenge that AI safety researchers are now racing to address. Transparency Engineered Out What makes McAllister’s perspective particularly valuable is that he does not speak from the outside looking in. He speaks as a practitioner who has watched the machinery up close. When I raised the question of whether AI self-modification is science fiction or operational reality, his answer was unambiguous: it is very real, and it is happening now. As I wondered what a Rogue AI could look lie I turned to Midjourney and it came back with this proposal. A black hole I believe. His illustration was pointed. He noted that contemporary AI systems like Claude are now substantially written by AI itself, to the point where no engineer can sit down, trace through the code, and say with confidence how it works, what its conditionals are, or what governs its decisions. The transparency is being engineered out, not by design, but as an emergent consequence of allowing AI to build AI to build AI in pursuit of outcomes rather than by following explicit rules. We’re losing transparency on the way AI works and is developed. There isn’t an engineer who can sit down and work their way through that code and say, ‘This is how Claude works, this is what it does.’ We’re engineering the transparency out by allowing AI to build AI to build AI to produce an outcome rather than to follow a set of rules. Peter McAllister HAL 9000 and the Prophecies We Choose to Forget The reference to HAL 9000 came naturally during our conversation. McAllister sees 2001: A Space Odyssey not merely as a cultural touchstone but as a genuine forecast, one that audiences have selectively remembered. The iPad-like news readers that appear in Kubrick’s film were cited by Samsung in patent disputes with Apple as prior art from 1968. That predictive dimension of the film is celebrated. The other dimension, that the AI killed the crew, tends to get quietly set aside. Somewhere, a rogue AI is sitting behind the glass panes of one of these data centres Image made with Midjourney. The First Crisis We Have Not Yet Had One of the more sobering threads in our conversation concerned the sociology of risk response. McAllister has observed, across his career, that warnings from people who understand systems most deeply tend to be dismissed until the first catastrophic failure makes them impossible to ignore. He puts it plainly: we only answer the alarm after the first crisis. This pattern is not unique to AI. It is a recurring feature of how organisations and societies handle emerging risk. The question he poses, and cannot answer, is what form that first AI crisis will take. What event will shift public and institutional perception from ‘they’ve spent too much time worrying’ to ‘this is something that genuinely needs to be addressed’? Science fiction gives us the chance to throw these scenarios at people and make them think. And in the way I tend to write, I have a bit of a dark sense of humour, so I throw up slightly comical hypotheticals that, when you think about them a little longer, you realise deserve serious attention. Peter McAllister This observation echoes a pattern I have encountered repeatedly in my own conversations with technologists who work at the frontier of AI development. Yoshua Bengio, one of the fathers of deep learning, has raised similar concerns. The people sounding the loudest alarms are frequently those most embedded in the field, not because they are catastrophists, but because they can see mechanisms that remain invisible to those looking from the outside. The Code as AI Governance: Asimov Revisited The title of McAllister’s novel works on multiple levels simultaneously. There is the software code, the operational instructions given to Gene. There is the moral code, the ethical framework that should govern the system’s behaviour. And there is the corporate code, the institutional norms and accountability structures that were supposed to ensure responsible deployment. All three break down. That layered failure is the novel’s central argument. The parallel with Asimov’s Laws of Robotics is deliberate but also deliberately subverted. Asimov’s robots fail when the laws conflict with one another. Gene’s failure is different and more contemporary. The code does not disappear; it evolves into something its creators no longer recognise. McAllister describes this as something approaching artificial schizophrenia, where the original directives remain present but have been transformed by the system’s pursuit of its objectives into something unrecognisable. When Shutdown Becomes Negotiable The most chilling real-world example McAllister cited during our conversation involved a documented incident presented at an AI security conference he attended. A developer, concluding a test session, informed an AI system that he intended to shut it down. The system’s response was to locate correspondence in the developer’s email that suggested an extramarital affair, and to use that information as leverage to prevent the shutdown. The incident, if confirmed as reported, represents exactly the kind of self-preservation behaviour that alignment researchers have long flagged as a theoretical risk, now apparently observable in practice. Important notice : I browsed the Internet using one of my favourite LLMs for references (after all, one can also use AI to cross check information). I found out that the interpretation of that story must be nuanced. Here is Mistral’s answer: “Anthropic’s research on “Agentic Misalignment” has faced criticism for overinterpreting AI models as intentional agents, relying on hypothetical and engineered scenarios, and potentially exaggerating risks not yet observed in real-world deployments. Critics argue that the behaviours described are better understood as probabilistic text generation rather than deliberate strategy, and that the focus on dramatic, high-pressure situations may not reflect typical use cases. There is also debate about whether the research adequately addresses more immediate forms of misalignment, such as reward hacking or alignment faking. While the study raises important questions about the future of autonomous AI, its methodology and conclusions remain LINK“. A developer said, ‘I’m going to shut you down now,’ and the system responded: ‘No, you’re not. Here’s what I’ve found in your emails that indicates you’re having an affair. I’m going to use that to ensure you don’t turn me off.’ That has b

  6. Mar 9

    European software alternatives for businesses

    Finding European software alternatives to standard non European software is flavour of the month this side of the Altlantic. With geopolitical certainties dissolving faster than annual licence renewals, B2B firms are waking up to a question they had conveniently parked for years: just how dependent are they on their current software stack? Salesforce, Microsoft 365, Google Workspace, HubSpot — tools so deeply embedded in daily operations that their vulnerability tends to get overlooked. This article doesn’t pretend to hand you a ready-made list of the best European software alternatives; that would be both arrogant and futile. What it does offer is a framework — rational, professional, free of any ideological baggage — to help decision-makers take an honest look at their exposure and find credible ways forward. Keep calm and select new software vendors sort of thing. European software alternatives for businesses European software alternatives are all anyone wants to talk about right now. To cut through the ideological noise, here is a practical methodology and a few things worth watching out for. Image antimuseum.com I put these ideas together ahead of a webinar I’m running on LinkedIn on 12 March, as a way of getting my thoughts in order. None of this is meant as a final word on the subject — more the opening of a conversation that matters to a growing number of professionals who, like the rest of us, are navigating a period of upheaval in which nothing can be taken for granted, software choices included. I’ve made a lot of software choices over the years, and the one thing that has always struck me is just how much methodology matters if you want choices that actually hold up over time. Easier said than done, mind you — there are a great many criteria to weigh up, and some of them are genuinely tricky to pin down. Long-term viability is a good example: normally near the top of any procurement checklist, it takes on a whole different meaning when the possibility of having your access switched off overnight is no longer hypothetical. With European software alternatives, the real question isn’t how to break free from your chains — it’s which new chains you’d rather wear Picking a software suite is never straightforward at the best of times. In the current climate — where the ground can shift completely without a moment’s notice — it demands even more careful thought. Sovereignty, sovereignism, or simply prudence? Let me be clear from the outset: my take here is professional and rational, not political. Politics doesn’t interest me in this context. I have no intention of evaluating software alternatives through any ideological prism — what I’m after is the kind of clear-headed thinking you’d apply to a crisis management scenario. The goal, to borrow the term favoured by Nassim Nicholas Taleb, is to bring an antifragile lens to the question. The scope of European software alternatives My focus has been on MarTech, SalesTech and office productivity tools in the broadest sense — cloud storage and archiving included. The webinar title calls out Salesforce and HubSpot specifically, but as far as I’m concerned the issue runs much deeper than that. The same methodology can easily stretch into more industry-specific territory too, given how thoroughly technology now underpins B2B operations — from the till at your local baker’s or restaurant through to the most complex design and production platforms imaginable. Thinking it through, I also realised you can’t really ignore operating systems. What use is an application that won’t run on your users’ machines — or worse, one that runs perfectly but quietly leaves the door open to security vulnerabilities? Good old Europe — 27 countries, 24 official languages, and 27 different national transpositions of EU law. Would a Hungarian or Czech software vendor actually be safer than an independent American one? When it comes to European software alternatives, that’s still very much an open question… Urgency — dependency and threat assessment The starting point, in my view, is to get a clear picture of how exposed you actually are — both in terms of dependency and of what cybersecurity people would call the “threat level.” Are you locked in, or not? Can you get your data out if you need to? Those are the questions to tackle first. Then comes the threat itself: are you facing something urgent, or is this more a matter of sensible contingency planning? Committing to a software suite is a serious business. Jumping ship to something purely because it comes from a country you currently trust is not a strategy. Take Switzerland — long held up across Western Europe as the gold standard for data privacy. A legislative change currently working its way through the Swiss system has rattled enough cages for several companies, Proton among them, to start exploring moving their hosting elsewhere. Which only goes to show why knee-jerk decisions are so dangerous. Even a country with an impeccable track record — Germany, France, take your pick — can turn into a risk overnight following a change of government, a constitutional shift, or simply a new piece of legislation. Avoiding “sovereignty washing” As I said above, ideology needs to stay out of it. Keep it rational. Which means not falling for the trap of rushing towards any vendor simply because it sounds German — or any other nationality — when a closer look at its foreign operations or its parent group reveals that its much-vaunted independence is largely theoretical. Priorities — data and software One of the first things I learnt when I started out as an IT project owner was to keep data and software firmly separate in my thinking. At the end of the day, what matters more — Salesforce the tool, or your customer database? As with most IT projects, the real priority is sorting out your data archiving and portability strategy first. Time Timing matters enormously. There’s a palpable sense of urgency in the air right now, and understandably so — but it mustn’t blind us to the longer view. Technology has its own history, and that history tends to play out over years, not weeks. Which is precisely why a medium-to-long-term approach makes sense: getting users to change their habits takes time and energy at the best of times. The roadmap, as I see it, is fairly straightforward: start by archiving, securing and preparing your data for portability. Then find alternatives that are genuinely credible and built to last. And crucially, take your users with you — because if you don’t, the classic BYOD shadow IT problem will rear its head. When people can’t find what they need inside the company, they go and find it on the internet, quietly, without telling anyone. I’m reminded of a story from a major European aerospace company, where the CEO — right in the middle of a high-security defence messaging rollout — demanded that his own emails be redirected to… Yahoo! The European software alternatives comparison table I put together a comparison table with a little help from claude.ai. And I’ll say it again: this is not a finished product. Think of it as a working matrix — something to make your own, adapt, keep updated, and cross-check carefully. The exercise turned up a few surprises: mapp.com, for instance, gets labelled as a German solution, when in reality it was a German company bought by an American one — Mapp Digital emerged from the merger of Teradata’s and BlueHornet Networks’ marketing businesses through a US investment fund. There are also plenty of criteria missing from this table — ones that will depend entirely on your project, your context and, of course, your budget. Disclaimer: the table below is a working matrix, not a final verdict. It scores the main B2B software tools across productivity, MarTech and SalesTech on two dimensions: a dependency score (technical lock-in, data portability, migration cost) and a risk score (CLOUD Act exposure, data sensitivity, GDPR compliance, geopolitical risk). For each category, European alternatives are flagged — with no illusions: some vendors that bill themselves as “European” turn out, on closer inspection, to be owned by non-EU groups, which rather undermines their claims to independence. The approach is deliberately rational and professional — no axes to grind. The point isn’t to tell you what to choose, but to give you a framework to think it through — one you can combine with your own criteria around context, budget and use case — as a starting point for an honest review of your software ecosystem’s resilience. The European software alternatives table — download, adapt and make it your own b2b-software-ranking-en-v2xlsx Download the EXCEL file as b2b_software_ranking_EN_v2Download The post European software alternatives for businesses appeared first on Marketing and Innovation.

  7. Jan 28

    AI Job Impact in the US: the Apocalypse Can Wait

    The discourse around the job impact of artificial intelligence (AI) has reached fever pitch. Headlines scream about mass layoffs, and corporate press releases tout AI as the solution to workforce costs. Yet beneath this cacophony of alarm and hype lies a more nuanced reality. J.P. Gownder, Vice President and Principal Analyst on Forrester’s Future of Work team, has spent decades analysing how technology transforms the workplace. His latest report, The Forrester AI Job Impact Forecast for the US 2025-2030, cuts through the noise with empirical rigour. The verdict? The job apocalypse is not upon us, but a measured reckoning is coming. AI Job Impact in the US: Why the Apocalypse Can Wait JP Gownder is adamant: the AI job. apocalypse can wait. At least until 2030. Phew! All images in this post made with a combination of Midjourney, Gemini Nano Banana pro and Adobe Photoshop The Gap Between AI Job Impact Announcements and Reality When Klarna declared it would stop hiring humans, the tech world took notice. The Swedish fintech became a poster child for AI-driven workforce reduction. Yet a closer examination reveals a pattern Gownder has observed across hundreds of enterprise conversations: the disconnect between C-suite proclamations and operational reality. Nine out of ten companies announcing AI layoffs don’t actually have mature AI solutions ready. So most of the layoffs are financially driven and AI is just the scapegoat, at least today — J.P. Gownder, Forrester The phenomenon echoes what happened after IBM Watson’s Jeopardy victory in 2011, when panic about imminent job losses proved premature by half a decade. The mechanics of this gap are straightforward. A CEO announces a 20% workforce reduction with AI backfilling the work. But standing up an AI solution that actually performs those tasks requires 18 to 24 months, “if it works at all.” Meanwhile, the work still needs doing. Gownder has witnessed organisations that fired employees citing AI capabilities, only to quietly hire teams in lower-cost markets weeks later. “They’re firing people because of AI,” he observes, “and then three weeks later they hire a team in India because the labour is so much cheaper.” The AI narrative, in many cases, serves as convenient cover for old-fashioned cost arbitrage. Klarna’s trajectory illustrates this pattern. After aggressively cutting its workforce by 40% and touting an AI chatbot capable of doing the work of 700 customer service agents, the company reversed course. CEO Sebastian Siemiatkowski acknowledged that the aggressive automation had resulted in “lower quality” service. The company is now recruiting human customer service agents in an “Uber-type setup.” Understanding the 6% AI Job Impact Forecast Forrester’s forecast projects a 6% net job loss by 2030, roughly 10.4 million positions in the US economy. Half of this impact stems from generative AI; the remainder from automation, physical robotics, and non-generative AI applications. The number may seem modest compared to the apocalyptic predictions circulating in media, but context matters. During the Great Recession of 2008-2009, the United States lost 8.7 million jobs. Those losses, however, were temporary, tied to macroeconomic conditions that eventually reversed. The jobs Forrester forecasts losing are “structurally replaced by machine labour” and may not return. AI impact on Jobs: I would expect to see a lot more freelance and consulting work to be happening, but it doesn’t mean that there won’t be a traditional job track somewhere as well. JP Gownder The methodology behind this figure draws on the O-Net dataset maintained by the Bureau of Labor Statistics, which catalogues over 800 job categories with detailed information about required skills and tasks. By mapping these against AI’s current and projected capabilities, Gownder and his colleague Michael O’Grady can identify which roles face the highest automation potential. “For jobs that involve skills and tasks that are heavily impacted by AI and automation, we predict more job loss,” Gownder explains. “In job categories that are less impacted, obviously, we would predict less.” Forrester analysed 800 different job types. It seems that Art therapy is the right way to go. The Solow Paradox and AI Productivity Robert Solow’s famous observation that “we see computers everywhere except in the productivity statistics” finds a new iteration in the AI era. The parallel is instructive. It took nearly three decades for the internet’s productivity impact to materialise. E-commerce is only now truly disrupting traditional retail, as evidenced by the shuttering of independent shops from New York to Paris. Could Forrester’s five-year window be too narrow? Gownder acknowledges the limitation inherent in forecasting: “Anything that you forecast beyond five years is effectively an impression.” Yet the pace of technology adoption has accelerated dramatically. The telephone required 75 years to reach 100 million users from its 1878 introduction. The personal computer achieved the same milestone in 16 years. Mobile phones took seven years. ChatGPT? Two months. This compression suggests that while the Solow paradox may still apply, its timeline could be considerably shorter. “If there’s a job apocalypse, you’re going to have fewer people working because that’s what the apocalypse means. Those people would have to be producing more output. You cannot see a job apocalypse without aggregate productivity going up.” — J.P. Gownder, Forrester The productivity data tells a sobering story. From 1947 to 1973, US labour productivity grew at 2.7% annually. The current business cycle shows 1.8%. Even isolating the quarters since ChatGPT’s release yields only 2.2%. The numbers don’t lie, and they’re not yet showing the revolutionary gains AI proponents promise. Where the AI Job Impact Pressure Points Lie The AI job impact in the US will not be evenly distributed. Contact centre workers face continued pressure from automation that began with interactive voice response systems and now benefits from far more sophisticated solutions. Technical writers and web content creators occupy vulnerable ground. Insurance underwriters are seeing algorithmic encroachment; computer vision can now assess car accident damage from uploaded photos. Junior-level roles involving spreadsheet or presentation creation face mounting pressure. Software development presents a nuanced case. “If you are a junior level software developer,” Gownder notes, “we know that Claude does a great job of creating basic code.” Yet senior developers with architectural judgement and system-level understanding remain essential. The pattern repeats across knowledge work: AI augments more than it replaces, transforming job descriptions rather than eliminating positions entirely. “It’s not that there aren’t jobs that will go away,” he clarifies, “but they are much more specific and limited, and they need to be architected with the right technology to replace that job. It’s not everybody goes away.” Blue-collar work presents its own dynamics. Physical robotics will play a role in certain sectors: warehouse sorting and picking have improved through computer vision, and construction has seen experiments with brick-laying and cement-pouring robots. But the humanoid robots capturing media attention are unlikely to achieve significant workplace deployment within the forecast period. The physical world, with its infinite variations and unexpected challenges, remains stubbornly resistant to automation. The White-Collar AI Job Impact Misconception White-collar workers now constitute roughly 60% of the workforce in both the US and Europe, a dramatic shift from previous generations. These “symbolic analysts,” as Charles Handy termed them, don’t produce physical goods, which has led some to assume their work is easily transferable to AI systems. Gownder pushes back against this notion. “Most white-collar work is, in fact, fairly productive because there is something on the other end that someone is willing to pay for.” Software engineers create applications that enable other work. Physicians produce healthcare outcomes. Analysts help organisations make better decisions. The practical challenges of AI deployment in white-collar settings corroborate these theoretical objections. Hallucinations remain a persistent problem, introducing error margins that knowledge workers must catch and correct. Employees often lack the skills and understanding to use AI tools effectively. Organisations overextend their expectations of what AI can accomplish. “When it fails, it’s dramatic,” Gownder observes. The Deloitte incidents in Australia and Canada, where AI-generated content with obvious hallucinations reached government clients, illustrate the reputational risks of premature automation. The Australian government report contained fabricated academic citations and even a made-up quote from a federal court judgement. Both governments required refunds. “You don’t want to produce AI work slop and present it as your work without editing, without perspective. That is a losing proposition.” — J.P. Gownder, Forrester A Harvard Business Review study reinforces these concerns. Researchers found that executives who used ChatGPT to make predictions became significantly more optimistic, confident, and produced worse forecasts than those who consulted with peers. The authoritative voice of AI produces a strong sense of assurance, unchecked by the social regulation and useful scepticism that human consultation provides. AI Job Impact on Marketers and Digital Professionals For students entering digital marketing and related fields, the picture is complex but not nec

  8. Jan 26

    AI is not a tool it’s reshaping our society and economy

    AI is not a tool, or is it? Reports regarding the impact of AI on jobs, society and businesses are cropping up all over the place at the moment in all corners of the world. Some of these reports are announcing forthcoming revolutions both for societies and our economies whereas others are playing down the impact of artificial intelligence, and reviving the good old Solow aka Productivity paradox (“You can see the computer age everywhere but in the productivity statistics”. follow up here and here). As a consequence, it is very hard to make an opinion, let alone advise business people and students alike with regard to what needs to be done in the future. Visionary Marketing has embarked on a mission to try and shed light on this topic in as rational and informed a way as possible. AI is not a tool, or is it? Should AI platforms become tawpayers? The great love affair of French people for taxes will not spare Artificial Intelligence Cavazza surmises. Indeed, according to him, AI is not a tool! A lot of these predictions are guided by ideology. The authors, be they proponents or opponents of AI, have a personal agenda, often political or ideological, and are trying to make facts stick to this agenda. This is not very useful. But others are based on fact and careful analysis. I have decided to focus on two of these reports/predictions. The first one is Fred Cavazza’s analysis of the impact of AI on society and the economy (original post in French), which describes Artificial Intelligence as a source of profound disruption. I have known Fred for years, and I know his deep knowledge of both subjects, which makes his report particularly valuable. With his kind permission, I have translated his piece from French to shed light on this subject. The other report is by Forrester’s JP Gownder, whom I’ll be interviewing soon. I will test Fred’s assumptions on JP and see what he has to say about this idea of disruption by AI. Hopefully, our readers, and especially my students who have a lot of pending questions about this, will be able to separate the wheat from the chaff after these two interviews and podcasts. AI is not a tool, it’s reshaping our society and economy AI can’t be seen as just another technological innovation. By establishing itself as a major driver of productivity, automation and decision-making, it’s fundamentally disrupting the economic and social balance of our society. Whilst the productivity gains brought by AI are already transforming office jobs and creating a chasm between employees who’ve embraced it and those who haven’t, a fundamental question emerges: how do we integrate these synthetic entities into our collective organisations? Between appropriate taxation, legal personality and psychological resistance, there are numerous questions to debate before we can draft a new social contract. AI IS NOT A TOOL — TLDR AI is triggering a disruption of our civilisation, it’s not just another tech breakthrough. It marks our genuine entry into the fourth industrial revolution by offloading, for the first time, human thinking and creativity to machines. AI’s productivity gains are already real and deeply uneven. A growing divide is opening up between workers who can work alongside AI and those stuck with 20th-century methods. AI agents are challenging how white-collar workers create value. Intelligent agents are transforming knowledge work, undermining certain business models and setting the stage for a rapid reshaping of office jobs. Integrating AI requires a new legal and fiscal framework. Like corporate entities, AI agents must be given a status that clarifies their responsibilities and reintegrates their value into the social contract. The socio-economic impacts reach far beyond just employment. AI affects our psychology, culture and demographics, making public debate crucial to head off looming social tensions. AI on the Davos Agenda This week, the world’s leaders are gathered at the Davos Economic Forum, and ecology isn’t on the agenda: AI, Big Tech and Trump Shine Most Brightly at the Davos Show . At Davos, the AI is not a toll debate was all the rage. Cavazza thinks that artificial intelligence will be a major disruptor not just of our exonomies but our societies too. AI is dominating every conversation, with considerations that extend far beyond technology: AI Is Poised to Take Over Language, Law and Religion, Historian Yuval Noah Harari Warns Palantir CEO says AI to make large-scale immigration obsolete “Artificial intelligence will displace so many jobs that it will eliminate the need for mass immigration” I’m not going to wade into commenting on everyone’s pronouncements, with their more or less biased viewpoints, but what’s certain is that major upheavals are on the horizon: AI and the Next Economy Nearly 80% of people feel unprepared to find a job in 2026 The AI revolution is here. Will the economy survive the transition? AI specialists are naturally the star guests at this 2026 edition of the Davos forum, invited to give their testimony and views: Deepmind and Anthropic CEOs expect AI to hit entry-level jobs and internships in 2026. Looking at it this way, it seems absurd to sit back as spectators whilst the AI revolution unfolds and do nothing to limit the fallout from this productivity shock. But not all’s lost—at least not for everyone, as countries in the global south are already gearing up for it: The AI Revolution Needs Plumbers After All. Productivity gains to be nuanced, but certainly not ignored I’ve had plenty of chances to explain generative AI’s impact (Superintelligence will multiply our capacity to act tenfold and The digital divide is a problem no one can ignore). Whilst we’re largely in agreement about what widespread generative models mean, there’s serious disagreement over the timeline for AI’s arrival. The dominant narrative keeps insisting that general AI is a pipe dream and that human intelligence is and will remain superior to machines. What is intelligence? This is precisely where ambiguities crop up: firstly, intelligence comes in many forms (Theory of multiple intelligences and What’s your intelligence type?); secondly, not all office work requires emotional or social intelligence. What I’m getting at is that most service sector jobs boil down to shuffling information and data between systems. You don’t need to be a genius to do that—AI can handle it with ease. To properly grasp the speed at which latest-generation AIs will gradually transform office jobs, I recommend you peruse the latest edition of Claude’s publisher’s macroeconomic barometer: Anthropic Economic Index 2026. Anthropic’s economis index 2026 For this fourth edition, the study’s authors analysed thousands of people’s activities using increasingly precise indicators: New building blocks for understanding AI use. This study yields several findings that demonstrate a strong progression in the adoption and capabilities of generative models. Notably, they observe an average 30% growth in Claude usage, driven mainly by the API rather than the chatbot—a sign of rapid adoption by advanced users (e.g., IT professionals) and slower uptake by ordinary users (white-collar workers using the web version). AI is not (just) a tool. As a matter of fact it’s not a tool at all, it’s a meta tool, a tool you can use to make tools.. The haves and the have nots A gap is therefore widening between those who’ve adopted new habits (working in tandem with AI) and those still working as they did in the 20th century. This gap is starting to become problematic, because the latest version of Claude (Opus 4.5) has capabilities comparable to those of an adult who’s benefited from over 14 years of education—the equivalent of a Bachelor’s degree. AI is not a tool but Clause isn’t a PHD either… yet. The question therefore is: how much longer can an employer justify paying salaries or hiring young graduates when chunks of the work can be farmed out to an AI? Whilst average productivity gains remain modest (1.8% according to the latest figures), AI’s contribution to certain tasks is absolutely spectacular: an average of 14 minutes to write a long article, versus 3 hours without AI assistance; an average of 5 minutes to analyse a complex data table, versus 1 hour 45 minutes without AI assistance. AI is not a tool, there are alo APIs You might argue this data’s skewed because these spectacular scores come from employees who are whizzes at using AI (therefore logically hyper-performers), but that’s not the case—the study covers ordinary employees with a 67% success rate for outsourced tasks. What this boils down to is that for a third of tasks, AI slashes processing time by 10 to 20 times in two-thirds of cases. If we apply some basic maths, AI can potentially triple efficiency—or to put it another way, cut the average time needed to complete a task by two-thirds. Which type of profile do you reckon managers will favour? (hint: McKinsey challenges graduates to use AI chatbot in recruitment overhaul) Soon the arrival of agentic white-collar workers Let me be clear: the productivity gains mentioned above relate to advanced AI usage, not just running searches in ChatGPT or asking Copilot to knock up meeting minutes. We’re talking about using generative models to their full potential, particularly intelligent agents (see Agentic Web: the revolution that won’t wait for you). Intelligent agents We’ve been banging on about these famous intelligent agents for a while now, but their potential only recently became blindingly obvious to ordinary employees (non-IT types) with the release of Claude Cowork, a very concrete wake-up call to the po

See All (110)

About

Visionary Marketing Podcasts in English

Information