Bertrand Duperrin
@duperrin.com
People/Organization/Tech professional and enthusiast.
Proud Parisian. Travel addict.
Blogging and Blueskying in 🇫🇷and 🇬🇧
Duperrin.com
Duperrin.com/english
Proud Parisian. Travel addict.
Blogging and Blueskying in 🇫🇷and 🇬🇧
Duperrin.com
Duperrin.com/english
AI streamlines standard HR tasks, but fails on human-sensitive cases. A coached employee quits after weeks of unnoticed distress, revealing AI can’t judge or escalate. The challenge isn’t adoption but discernment: set clear limits, keep human responsibility.
The inconvenient truth about AI in 2026: Humans are the complicated part
Four experts weigh in on 2026 for HR leaders: the challenge isn’t adopting AI, it’s knowing when not to use it.
hrexecutive.com
November 19, 2025 at 12:58 PM
AI streamlines standard HR tasks, but fails on human-sensitive cases. A coached employee quits after weeks of unnoticed distress, revealing AI can’t judge or escalate. The challenge isn’t adoption but discernment: set clear limits, keep human responsibility.
“ #AI #skills ” is a vague label. Real competence isn’t tool use but the ability to drive change: align intentions with uses, redesign processes, and apply human judgment where machines can’t.
AI skills: a catch-all term for a major misunderstanding
Today, everyone is talking about “AI skills.” They are everywhere: in job descriptions, training plans, and even in job interviews, where they hav
www.duperrin.com
November 19, 2025 at 10:12 AM
WSJ launches a CPO Council with Workhuman, arguing HR must lead AI adoption as corporate value shifts to people-centric intangibles. Alan Murray and KeyAnna Schmiedl call for redesigning work, breaking silos, and strengthening IT–HR alignment, with clear governance to turn AI into measurable impact.
AI is a people challenge, not a tech one: Here’s how to redesign work for the better, according to Workhuman & WSJ Leadership Institute
In exclusive interviews at Workhuman Forum on Day One of UNLEASH World, KeyAnna Schmiedel, CHXO at Workhuman, and Alan Murray, President of the WSJ Leadership Institute, explore the crucial role that ...
www.unleash.ai
November 19, 2025 at 7:22 AM
WSJ launches a CPO Council with Workhuman, arguing HR must lead AI adoption as corporate value shifts to people-centric intangibles. Alan Murray and KeyAnna Schmiedl call for redesigning work, breaking silos, and strengthening IT–HR alignment, with clear governance to turn AI into measurable impact.
La prétendue obsolescence des hard skills repose sur une mauvaise interprétation d’un chiffre du WEF. Les compétences évoluent plus qu’elles ne disparaissent. Ce discours nourrit anxiété et discriminations alors que l’IA renforce la valeur d’un socle de connaissances solides.
Pourquoi faut-il en finir avec l’obsolescence rapide des hard skills
Malgré un article à succès au Harvard Business Review qui démonte ce bullshit ou encore une interview pour Welcome To The Jungle, la croyance de l’obsolescence rapide des hard skills continue de sévir...
www.linkedin.com
November 18, 2025 at 12:11 PM
La prétendue obsolescence des hard skills repose sur une mauvaise interprétation d’un chiffre du WEF. Les compétences évoluent plus qu’elles ne disparaissent. Ce discours nourrit anxiété et discriminations alors que l’IA renforce la valeur d’un socle de connaissances solides.
Sutherland takes a gradual, controlled approach to #AI in #HR, boosting productivity without replacing human interaction. AI improves sourcing and screening, while people lead key steps to ensure trust and authenticity. An AI Academy supports skills and governance secures tools and data.
Prioritizing AI integration in 2026? 6 questions to build a framework
The HR executive of digital transformation firm Sutherland shares the comprehensive framework his org is following for AI integration.
hrexecutive.com
November 18, 2025 at 6:12 AM
AI agents left alone. Journalist Evan Ratliff built a fake startup, HurumoAI, run only by agents to test the “one-person billion-dollar company” idea. Results: busy work, little impact, and even a fake offsite that burned credits. A prototype emerged, but agents still fail at basic tasks.
Company Run Almost Entirely by AI-Generated Employees Descends Into Chaos
Journalist Evan Ratliff populated a fictional tech startup, dubbed HurumoAI, exclusively with AI agents to see what would happen.
futurism.com
November 17, 2025 at 5:58 PM
AI agents left alone. Journalist Evan Ratliff built a fake startup, HurumoAI, run only by agents to test the “one-person billion-dollar company” idea. Results: busy work, little impact, and even a fake offsite that burned credits. A prototype emerged, but agents still fail at basic tasks.
AI sparks frontline anxiety: many fear replacement by automation or AI-skilled coworkers, made worse by little clarity from employers and limited digital access. Trust requires training, simple use cases, supportive managers, and clean, connected HR data to enable fair adoption.
Empowering the Frontline Workforce with AI and Human-Centric Leadership
Insights from a closed-door conversation with senior HR leaders at UNLEASH World 2025, hosted by UKG.
www.unleash.ai
November 16, 2025 at 3:04 PM
AI sparks frontline anxiety: many fear replacement by automation or AI-skilled coworkers, made worse by little clarity from employers and limited digital access. Trust requires training, simple use cases, supportive managers, and clean, connected HR data to enable fair adoption.
Digital twins now model individuals: platforms like Viven.ai create active replicas from emails, meetings and documents to access expertise and ease coordination. Strong potential for knowledge flow, but data ownership, privacy and governance raise significant concerns.
Arriving Now..... The Digital Twin.
Introducing the "digital twin," a new way to leverage AI for collaboration, productivity, teamwork, and collective business intelligence.
joshbersin.com
November 15, 2025 at 12:55 PM
Digital twins now model individuals: platforms like Viven.ai create active replicas from emails, meetings and documents to access expertise and ease coordination. Strong potential for knowledge flow, but data ownership, privacy and governance raise significant concerns.
#AI #adoption in business reveals a gap between ambition and real change. Most firms claim to use AI but lack redesigned processes or governance. True transformation, where AI reshapes work and decision-making, remains rare. Mature firms treat AI as organizational design, not just technology.
Adoption of AI in the workplace : current situation
For two years now, everyone has been talking about artificial intelligence in business, which seems logical given the promise of this technology. But in reality
www.duperrin.com
November 14, 2025 at 11:56 AM
AI is booming on billions but still lacks a viable model. Massive spending on data, chips, and talent drives progress, yet profits lag. High inference costs strain flat-fee plans. Nvidia wins, OpenAI burns cash. The race risks a price war before finding sustainable value capture.
AI Companies Don’t Have a Profitable Business Model. Does That Matter?
The generative AI boom is fueled by staggering investments (including OpenAI’s multibillion-dollar chip deals), but for many companies, profitability as a result of these investments has remained elus...
hbr.org
November 14, 2025 at 9:13 AM
AI is booming on billions but still lacks a viable model. Massive spending on data, chips, and talent drives progress, yet profits lag. High inference costs strain flat-fee plans. Nvidia wins, OpenAI burns cash. The race risks a price war before finding sustainable value capture.
#Knowledge is no longer scarce but abundant and automated. Drucker’s “knowledge worker” gives way to the “value creator”: one who transforms shared information into tangible outcomes through synthesis, trust, and adaptive learning. The new leadership designs conditions, not controls.
Beyond the Knowledge Worker: The Rise of the Value Creator By Mark Béliczky & Hunter Hastings - Global Peter Drucker Forum BLOG
Peter Drucker did not simply name the knowledge economy — he equipped us to navigate it. His central insight was profound: when knowledge became the primary resource of advanced economies, the knowled...
www.druckerforum.org
November 13, 2025 at 10:54 AM
#Knowledge is no longer scarce but abundant and automated. Drucker’s “knowledge worker” gives way to the “value creator”: one who transforms shared information into tangible outcomes through synthesis, trust, and adaptive learning. The new leadership designs conditions, not controls.
Leaders only see 4% of their organization’s problems, a gap first shown by Sidney Yoshida. Despite data abundance, structural silos and fear still block truth from the top. Real management means designing trust and friction zones where signals surface before automation widens the blindness.
Learn about Yoshida's iceberg of ignorance, or what management refuses to see.
In the life of organizations, there are truths that we prefer to ignore, until the day they brutally remind us of their existence. The image of an iceberg illus
www.duperrin.com
November 13, 2025 at 6:18 AM
Leaders only see 4% of their organization’s problems, a gap first shown by Sidney Yoshida. Despite data abundance, structural silos and fear still block truth from the top. Real management means designing trust and friction zones where signals surface before automation widens the blindness.
Most firms see little real gain from #AI. The barriers are not technical but organizational, rooted in people, processes, and power. Fear, outdated workflows, and disrupted hierarchies slow adoption. Only firms aligning skills, incentives, and design around AI achieve lasting value.
Overcoming the Organizational Barriers to AI Adoption
Most firms struggle to capture real value from AI not because the technology fails—but because their people, processes, and politics do. Survey data and case studies demonstrate how fear of replacemen...
hbr.org
November 12, 2025 at 3:18 PM
Most firms see little real gain from #AI. The barriers are not technical but organizational, rooted in people, processes, and power. Fear, outdated workflows, and disrupted hierarchies slow adoption. Only firms aligning skills, incentives, and design around AI achieve lasting value.
For #Airbus VP #HR Digital & AI Vincent Dupuis, HR must design the future of work by redefining roles and fostering critical thinking. #AI shouldn’t reduce humans to validators. Airbus aligns tech use with its mission, placing ethics, judgment, and adaptability at the core.
Airbus: HR leaders’ role is to be the ‘architects of the future of work’
UNLEASH sat down with Airbus’ Vincent Dupuis during this year’s UNLEASH World to get the inside track on how the aerospace giant is building AI into its people strategy.
www.unleash.ai
November 12, 2025 at 1:15 PM
Enterprise design restores coherence by reconnecting strategy, operations, and lived experience. Rather than optimizing processes, it aligns what a company wants to be with how it acts. Clarity becomes a competitive advantage when technology reshapes structures faster than management can adapt.
A poorly designed enterprise is illegible and incomprehensible to employees and customers
I have seen many businesses where competent people work with dedication and conviction, but where no one seems to have any idea what they are doing. They have l
www.duperrin.com
November 12, 2025 at 9:00 AM
Enterprise design restores coherence by reconnecting strategy, operations, and lived experience. Rather than optimizing processes, it aligns what a company wants to be with how it acts. Clarity becomes a competitive advantage when technology reshapes structures faster than management can adapt.
AI adoption isn’t enough. Most AI projects fail not from poor execution but from wrong goals. Success lies in using AI to redesign how value is created and captured, as Fujifilm, Shutterstock, or Walmart did, turning technology from an add-on into a lever for business reinvention.
What Executives Get Wrong About AI
Most companies fail to realize that true advantage with AI and other transformative technologies comes not from mere adoption but from reinvention—rethinking business models, value creation, and organ...
hbr.org
November 11, 2025 at 12:42 PM
AI adoption isn’t enough. Most AI projects fail not from poor execution but from wrong goals. Success lies in using AI to redesign how value is created and captured, as Fujifilm, Shutterstock, or Walmart did, turning technology from an add-on into a lever for business reinvention.
Greek and Roman models offer two timeless lenses for organizational design: the Greek model values clarity, defined roles, and control; the Roman model favors integration, flow, and adaptability. The real strength lies in combining both structural resilience with operational agility.
Turning Classical Design Thinking into a Human-Centred Business Playbook By Nick Hixson - Global Peter Drucker Forum BLOG
Business expansion presents organisations with dilemmas as old as commerce itself: how to scale without sacrificing human connection, and how to balance the demands of stability with the imperatives o...
www.druckerforum.org
November 11, 2025 at 9:57 AM
Greek and Roman models offer two timeless lenses for organizational design: the Greek model values clarity, defined roles, and control; the Roman model favors integration, flow, and adaptability. The real strength lies in combining both structural resilience with operational agility.
Geoffrey Hinton warns that massive AI investments, still unprofitable, can only pay off through large-scale automation replacing human labor. Without structural change, productivity gains will enrich infrastructure owners, not workers, pushing capitalism toward jobless growth.
The AI Industry Can't Profit Unless It Replaces Human Jobs, Warns Man Who Helped Create It
Geoffrey Hinton is warning that the AI industry is fundamentally built on the total destruction of human labor.
futurism.com
November 10, 2025 at 3:55 PM
Geoffrey Hinton warns that massive AI investments, still unprofitable, can only pay off through large-scale automation replacing human labor. Without structural change, productivity gains will enrich infrastructure owners, not workers, pushing capitalism toward jobless growth.
Sam Altman dodged questions about how OpenAI will fund its rumored $1.4T ambitions on $13B revenue, hinting at vague growth and future ventures. Days later, its CFO called AI a “national strategic asset,” signaling a shift toward taxpayer-backed funding under the guise of U.S.–China rivalry.
OpenAI probably can’t make ends meet. That’s where you come in.
If you thought the 2008 bank bailout was bad, wait til you see the 2026 AI bailout
garymarcus.substack.com
November 10, 2025 at 7:57 AM
Sam Altman dodged questions about how OpenAI will fund its rumored $1.4T ambitions on $13B revenue, hinting at vague growth and future ventures. Days later, its CFO called AI a “national strategic asset,” signaling a shift toward taxpayer-backed funding under the guise of U.S.–China rivalry.
#AI is turning the #web into a background system. As agents replace users, navigation gives way to delegation, content is consumed without credit, and value shifts to intermediaries. The web becomes a closed data layer : efficient, automated, but stripped of openness and diversity.
Dystopia: what if artificial intelligence broke the web and the Internet?
In my "dystopias" series, I wanted to try to describe fictional worlds where certain trends in today's world had been taken to extremes. On reflection, I wonder
www.duperrin.com
November 9, 2025 at 12:25 PM
Six barriers hinder AI skills development in the UK: unclear definitions of “AI skills”, low digital literacy, fragmented training systems, outdated curricula, unstable funding, and weak employer demand. Together, they limit access, consistency, and alignment between learning and labour needs.
AI skills development barriers
The UK government has outlined six barriers to AI skills development based on new research and analysis
www.processexcellencenetwork.com
November 9, 2025 at 8:06 AM
Six barriers hinder AI skills development in the UK: unclear definitions of “AI skills”, low digital literacy, fragmented training systems, outdated curricula, unstable funding, and weak employer demand. Together, they limit access, consistency, and alignment between learning and labour needs.
Trust in leadership drives performance, loyalty, and innovation, yet few firms measure it rigorously. Tools like the Leadership Trust Index show it can be tracked and improved. Regular measurement turns trust from an abstract virtue into a strategic asset and early-warning system for leaders.
If Trust Is So Important, Why Aren’t We Measuring It?
As stakeholder expectations grow and organizational life becomes more complex—especially with remote work, social accountability, and generational shifts—leadership trust is under more scrutiny than e...
hbr.org
November 8, 2025 at 1:55 PM
Trust in leadership drives performance, loyalty, and innovation, yet few firms measure it rigorously. Tools like the Leadership Trust Index show it can be tracked and improved. Regular measurement turns trust from an abstract virtue into a strategic asset and early-warning system for leaders.
Tech stacks built for speed can hollow out service quality. Klarna’s automation gains cut human touch, self-checkouts hurt experience, and even Southwest shows that smooth systems can fail under stress. Enterprise design restores intent by aligning identity, operations, and service.
Taking back control of enterprise design: intention before tools
For a long time, it was believed that efficiency would come from tools, and sometimes even from tools alone, so we piled up platforms, workflows, and assistants
www.duperrin.com
November 8, 2025 at 8:52 AM
Tech stacks built for speed can hollow out service quality. Klarna’s automation gains cut human touch, self-checkouts hurt experience, and even Southwest shows that smooth systems can fail under stress. Enterprise design restores intent by aligning identity, operations, and service.
Hidden “shadow work” drains $1.7 trillion a year. Employees spend seven hours weekly on invisible admin tasks, plus 11 minutes to refocus after each. Nearly half link it to burnout, yet only 7% have integrated tools. Reducing it is key to engagement, retention, and meaningful work.
Companies waste $1.7 trillion every year on unproductive ‘shadow work’, finds Perk
Shadow work includes tasks like filing expenses, booking travel and chasing approval; it is time consuming, and distracting for employees, and has a knock on effect on organizational productivity. Tha...
www.unleash.ai
November 7, 2025 at 11:01 AM
Hidden “shadow work” drains $1.7 trillion a year. Employees spend seven hours weekly on invisible admin tasks, plus 11 minutes to refocus after each. Nearly half link it to burnout, yet only 7% have integrated tools. Reducing it is key to engagement, retention, and meaningful work.
Agentic AI is moving from promise to results. A few pioneers now use agents that reason, decide, and act across customer journeys—optimizing pricing, personalizing campaigns, and anticipating needs. Real gains come from redesigning workflows, not adding tools, and training humans to work with AI.
www.mckinsey.com
November 7, 2025 at 7:52 AM
Agentic AI is moving from promise to results. A few pioneers now use agents that reason, decide, and act across customer journeys—optimizing pricing, personalizing campaigns, and anticipating needs. Real gains come from redesigning workflows, not adding tools, and training humans to work with AI.