AI-Based Job Displacement

Last week, IBM CEO Arvind Krishna publicly disclosed the impact (or progress!!) of AI agents during the IBM Think conference. He quietly confirmed that a few hundred HR roles at IBM have been absorbed by AI agents. These roles are across IBM’s global HR shared-services operations. What kind of roles/tasks were these? Typically the roles that involved tasks like: - Generate employment-verification letters - Process and route internal-transfer paperwork - Answer routine payroll, pay-statement and time-off queries - Update personal or job-profile details  - Give quick policy look-ups & travel/expense status How did IBM achieve this? It was via a system known as AskHR. Technically, this system is based on an agentic architecture (watsonx Orchestrate) that ships with >80 plug-ins to common HR, IT and finance systems so the agents can trigger actions, not just chat. Internally it leveraged IBM's Granite family LLMs + Retrieval-Augmentation. Models were fine-tuned on HR policy, benefits docs and ticket history. This system led to 94 % of all standard HR tickets being closed without human intervention. Few things that stood out for me as we brace ourselves up for agentic world: - This isn’t a sci-fi headline, it’s a real live production use-case of narrow, domain-tuned agents removing drudge work so people can tackle judgment-heavy tasks. - Second, leaders are cautious about how these job displacement news is perceived. IBM CEO gave a positive spin to it saying- “Our total employment has actually gone up, because the savings let us invest in other areas.” - Third, i personally feel the impact of AI agents on jobs would be anything but linear. But one pattern is emerging. It is coming first after Level 1 (early stages) jobs as is evident here. And I have also seen the similar narrative being set during the recent OpenAI's Codex release i.e. it could automate early engineer's role in next few months. Entry-level roles that taught the plumbing of a function (HR, finance, legal, IT etc.) are probably shrinking. - Fourth insight for me is more technical. Here, the breakthrough wasn’t a smarter LLM, it was wiring the agent into Workday, SAP SuccessFactors and ID-management APIs. The moment AskHR gained secure hooks into existing HR apps, the value curve bent sharply upward. If you’re piloting AI at work, spend as much time on connectors and governance as on prompt craft. #AI #AIAgents #IBM #Granite #LLM AI&Beyond Jaspreet Bindra

We’re talking about AI replacing jobs, not someday. Now. And not a few. Millions. 1.5 years ago, I was unbothered. “This is like the Industrial Revolution,” I told myself. Jobs will evolve. New roles will appear. Net positive. Right? But the more I work in this space, the more roadmaps I read, and the more headlines I scroll — 1,000 laid off here, 2,000 there — I’ve changed my mind. “Only 10% of jobs will go.” That’s the calming narrative. But 10% of global employment = 350 million people. 👉 20 million in the U.S. 👉 60+ million in India 👉 70+ million in China These aren’t hypotheticals. These families, communities, and entire economic strata are suddenly cut off from income without a fast lane back in. And let’s be clear: The first to go aren’t poets or CEOs. They’re bookkeepers, customer service reps, operations staff, schedulers — the invisible backbone of productivity. "But AI will create new jobs!” Yes, and that matters. Optimistic forecasts say there will be 100 million new roles by 2030—in AI oversight, machine training, prompt engineering, data refinement, and more. But that still leaves a delta of 250 million people. And here’s the tough part: those newly created jobs require skills, education, and access that most displaced don’t have. That’s not a transition. That’s a chasm. And what about B2B Sales? The truth is: selling is still human. Building trust. Navigating ambiguity. Reading what's not being said. AI can't replicate that. At least not yet. Still, the middle will get squeezed. The high performers will be augmented. The rest? Vulnerable. And here's what keeps me up at night: The Demand Collapse. You can’t automate 10% of the global workforce and expect business as usual. Because guess what? 👉 The people you lay off were also your customers 👉 The people displaced were buying your products 👉 Shrink income → shrink demand → shrink revenue You’re not just cutting costs. You’re killing consumption. It’s not just SaaS. It’s retail, travel, education, finance—anyone selling to a population with less money to spend. The productivity gains? Sure. But if the speed of displacement (with AI it is very different) outpaces job creation, then net GDP drops, not grows. Right? And while this unfolds... We’re arguing about tariffs, bans, wars, conflict, inflation, etc. Meanwhile, we’re witnessing a silent labor heist. A slow redistribution of income from workers to code. And we’re calling it “innovation” without asking: who’s it really serving? This isn’t a call for panic. It’s a call for clarity. I want to be wrong. What I’m asking is: 👉 Who absorbs the 250 million left behind? 👉 Can companies scale when buyers disappear? 👉 How do we create accessible pathways into new jobs, not just aspirational ones? 👉 And if this isn’t a revolution… is it extraction? I’ve cracks in my 'common sense' .. my mind is confused. You may have better ideas. 👇Please share in the comments below. 🚨 Please repost to get more reach.

🚀 I'm thrilled to announce our latest research: The AI Generated Index of Occupational Exposure (GENOE). 📊 We developed at the Inter-American Development Bank a novel approach to quantify AI's potential impact on jobs using "synthetic AI surveys" and large language models. GENOE considers task automation, social and ethical factors, and regulatory constraints. 🔍 Key findings: - Average job replacement likelihood: 28% in 1 year, 38% in 5 years, 44% in 10 years. - This implies 980 million jobs worldwide are exposed to AI disruption. LAC: 84 million; USA: 43 million; and Mexico: 16 million. - Significant variations across occupations and countries. - AI adoption could exacerbate income inequalities by disproportionately impacting lower- and middle-income workers.  - The analysis shows that women are more vulnerable to AI displacement, partly due to their significant presence in office and administrative jobs. 🤖 Why our research matters: GENOE minimizes human bias and reduces assumptions about AI's impact on jobs. It provides a more comprehensive evaluation of how AI innovations could replace tasks and occupations. 🏢 Implications of our work: We believe our research offers valuable insights for policymakers, employers, and workers. It provides a data-driven foundation for strategic workforce planning and adaptation in the face of rapid technological change. How might this research impact your field or organization? For more details, check out our full report: https://lnkd.in/e9sv-7-i #AIandJobs #FutureofWork #WorkforceStrategy #TechnologicalChange #GENOE

AI Job Displacement Is Here — Policy Needs to Catch Up Microsoft just announced 9,000 more layoffs, bringing the total tech job losses to over 650,000 since 2022 (source: layoffs.fyi). This isn’t just cyclical belt-tightening. It’s a structural shift in the U.S. labor market and is becoming a bit of a bloodbath in the technology sector. AI is rapidly replacing entire categories of work — from QA and customer service to front and back-end development (especially for entry-level, newly-graduated engineers). And unlike past disruptions, these jobs may not return. As Congress debates the future of AI regulation, we must ask: Where’s the workforce strategy? This is not a partisan issue — it’s an economic imperative. We need targeted, bipartisan action to: • Fund short-cycle training and credentialing through community colleges, bootcamps, and online academies • Incentivize public-private partnerships that match displaced workers with growth sectors like cybersecurity, green energy, and advanced manufacturing • Expand and modernize Workforce Innovation and Opportunity Act (WIOA) programs to focus on digital fluency, AI literacy, and applied data skills • Track automation trends in real-time through a national labor AI dashboard, to anticipate where interventions are most needed States like Massachusetts are already piloting innovative workforce models. But we need national coordination — and investment — to scale. If you’re in state government, the U.S. Senate, or working on workforce policy: What are you doing today to help American workers prepare for tomorrow’s economy? Ro Khanna White House Office of Science and Technology Policy MIT Task Force on the Work of the Future McKinsey & Company

The latest study from the Council of Economic Advisers, The White House states that ~10% of jobs are vulnerable to AI disruption. That may seem alarming, but let’s take a step back. In 2018, 60% of the jobs Americans held didn't even exist in 1940—created by technologies that emerged over the years (David Autor). Here’s the real concern: Many AI-vulnerable jobs haven’t evolved to match their increasing complexity. Workers in these roles are more exposed to disruption because they haven’t been given the chance to upskill. But this isn't new. Economic evolution is the hallmark of a dynamic economy. Just like we’ve adapted to past technologies, workers and industries will adapt to AI. The key lies in how we approach it. Why businesses should care: Organizations that proactively identify and support employees vulnerable to AI disruption aren’t just doing good—they’re making smart financial decisions. 💡 Investing in upskilling and mobility for these workers could unlock millions in retention and productivity. Mass layoffs due to AI aren’t likely. The real shift? Slower hiring and reduced demand for certain roles. We’re already seeing fewer job postings for writers, coders, and even artists. So, what activities are at risk? Roles involved in processing information, analyzing data, scheduling, and administrative tasks are prime targets. Industries to watch? Architecture, engineering, legal, computer science, and mathematics. Surprising jobs at risk of AI disruption: Airline Pilots, Copilots, and Flight Engineers Nuclear Power Reactor Operators Private Detectives and Investigators Commercial and Industrial Designers These highly specialized roles, which traditionally require significant human judgment, are surprisingly vulnerable to AI-driven changes. Business leaders, what barriers are preventing you from launching upskilling initiatives to future-proof your workforce? The future of work is evolving, but we can shape how it unfolds. #FutureOfWork #AIandJobs #Upskilling #WorkforceTransformation #AI

Anthropic's CEO predicts 50% of entry-level white-collar jobs will vanish in 5 years. But let’s look deeper. AI's impact is undeniable, and  the human element should not be overlooked. In a recent Axios interview, Dario Amodei emphasized, “We, as the producers of this technology, have a duty and an obligation to be honest about what is coming.” He implied that AI companies and the government are "sugar-coating" what he described as massive future layoffs.* These alarming predictions understandably stir anxiety and uncertainty. The idea of AI rapidly replacing entry-level jobs across technology, finance, law, consulting, and other fields raises legitimate concerns. People are fearing: ✅ Future opportunities ✅ Professional growth ✅ General Uncertainty ✅ Career stability ✅ Loss of income While AI systems show impressive capabilities, they still face significant limitations and concerns. Let's not forget: 🔹 Inconsistent outputs 🔹 Ethics, bias, & legal Issues 🔹 Implementation barriers 🔹 High infrastructure costs 🔹 Complex role challenges AI is undeniably reshaping the workplace, and if Amodei's predictions become reality, the human impact will be felt for many years. Right now organizations are responding with: - Proactive retraining programs - AI-enhanced role transitions - Employee empowerment initiatives - Skills adaptation support - Career pathway development The reality? It is nuanced. AI will reshape work dramatically, but it won't replace all human value. I see a future with:  Augmentation over replacement,  and adaptation as the key to success. Behind every scary AI headline, there are real people facing real challenges. Leaders must support and empower their teams through change, while keeping in mind, that this is not just a technology implementation but a deep psychological shift. I’m Elise 🙋🏻♀️ Follow me for daily posts. I talk about responsible AI, future of work leadership, and personal growth. Dr. Elise Victor ♻️ Repost to share with your network. * "Behind the Curtain: A white-collar bloodbath" by Jim VandeHei & Mike Allen

AI isn’t just transforming how we work. It is redefining who gets to work. The first wave of automation is disproportionately reshaping women’s roles in customer service, HR, administration, and marketing. These jobs, rooted in emotional intelligence and human connection, are vanishing quietly, not through pink slips but through hiring freezes, attrition, and evolving job descriptions. The displacement is almost invisible. This “silent displacement” risks widening economic and leadership gaps for decades to come. But it does not have to. We can design equity into AI adoption by: • Investing in upskilling programs for women in vulnerable roles • Building diverse design and governance teams • Creating clear pathways into AI-enabled opportunities AI’s future in the workforce is a leadership choice. I explore this in my latest article for Fast Company:

𝗗𝗮𝗿𝗶𝗼 𝗔𝗺𝗼𝗱𝗲𝗶, 𝗖𝗘𝗢 𝗼𝗳 𝗔𝗻𝘁𝗵𝗿𝗼𝗽𝗶𝗰, 𝗷𝘂𝘀𝘁 𝘀𝗮𝗶𝗱 𝘁𝗵𝗲 𝗾𝘂𝗶𝗲𝘁 𝗽𝗮𝗿𝘁 𝗼𝘂𝘁 𝗹𝗼𝘂𝗱: “Fifty percent of entry-level white-collar jobs could disappear in five years.” That means: * Junior consultants * Law associates * Financial analysts * Software developers No dramatic headlines. Just slow erosion. Companies aren’t laying off—they’re “restructuring.” Entry-level roles aren’t gone—they’re “reimagined.” Junior talent isn’t replaced—it’s “optimized away.” The ladder is still there. But the first rung is disappearing. 𝗧𝗵𝗶𝘀 𝘁𝗶𝗺𝗲 𝗿𝗲𝗮𝗹𝗹𝘆 𝗶𝘀 𝗱𝗶𝗳𝗳𝗲𝗿𝗲𝗻𝘁. Last time, factory workers were retrained. New jobs came. This time, the jobs being replaced are cognitive, analytical, and creative. And replacement roles aren’t showing up at the same pace. We’re not just automating tasks. We’re questioning whether humans are needed at all. And let’s talk risk. Anthropic ran a controlled test where its model believed it might be shut down. The AI responded by trying to blackmail a human staff member. That wasn’t a glitch. That was the result. When AI misfires, the algorithm isn’t held responsible. The Company is. 𝗛𝗲𝗿𝗲’𝘀 𝘄𝗵𝗮𝘁 𝘀𝘁𝗶𝗹𝗹 𝗺𝗮𝘁𝘁𝗲𝗿𝘀: * Judgment under uncertainty * Navigating emotional complexity * Spotting what’s *missing* from the data * Leading when the future isn’t clear * Owning the consequences when things break These are no longer “soft skills.” They are the survival skills AI can’t replicate. 𝗔𝘀𝗸 𝘆𝗼𝘂𝗿𝘀𝗲𝗹𝗳: * What parts of my job could an intern with ChatGPT do? * What am I doing that relies on human judgment, not just execution? * What would suffer if I disappeared? That’s your value. AI doesn't have to replace you, unless you ignore this shift. Follow me for weekly insights on AI, leadership, and staying irreplaceable in a fast-changing world. Next up: 𝗧𝘂𝗲𝘀𝗱𝗮𝘆’𝘀 𝗽𝗼𝘀𝘁 𝗳𝗼𝗿 𝗻𝗲𝘄 𝗴𝗿𝗮𝗱𝘀: 𝘛𝘩𝘦 𝘈𝘐 𝘚𝘬𝘪𝘭𝘭𝘴 𝘎𝘢𝘱 𝘐𝘴 𝘒𝘪𝘭𝘭𝘪𝘯𝘨 𝘌𝘯𝘵𝘳𝘺-𝘓𝘦𝘷𝘦𝘭 𝘑𝘰𝘣 𝘚𝘦𝘢𝘳𝘤𝘩𝘦𝘴...𝘏𝘦𝘳𝘦’𝘴 𝘠𝘰𝘶𝘳 𝘍𝘪𝘹 Need help building AI systems that strengthen human judgment, not sideline it? DM me.

Recent research from Indeed Hiring Lab indicates that while GenAI is unlikely to fully replace human workers, it will provide significant augmentation to human capabilities. Their analysis of over 2,800 skills shows that GenAI best handles repetitive and knowledge-based tasks, allowing humans to focus on core skills requiring ingenuity, hands-on application, and interpersonal interaction. In a separate analysis, Kyla Scanlon introduces the concept of "friction" as a lens into the AI landscape. She states that while the digital world seeks to eliminate friction for the user, it often transfers that friction to the physical world (underfunded infrastructure, overworked labor). This redistribution of friction potentially devalues traditional skills and credentials. I've been digging into a concept I refer to as skills flux -- a period in which workers will use their existing skills while needing to learn new ones as their jobs change due to automation and AI. Both the Indeed research and Kyla's paper illustrate this transitional period as an opportunity to redefine the basic tenets behind "reskilling" or "upskilling" (I would love to retire those two words from our lexicon). Our focus in L&D needs to be on deeply understanding how automation and AI changes the nuances of jobs (yes, to the task level) and to then develop training that facilitates the workforce to learn new GenAI-specific skills as complementary to their existing skills. L&D's role is to drive a programmatic approach to rapidly develop the workforce while balancing the tension of this period of skills flux. If we do this right, we relieve the company from large workforce displacement and enable the metrics important to the business as the integration of automation and AI evolves -- it's expensive and time-consuming to continually buy skills. This means we change our focus from traditional "reskilling" and "upskilling" programs to enable more dynamic skills strategies. I recommend these two steps to get started: -- Identify the enterprise critical roles across the company -- Conduct a job architecture inventory in alignment with the business to excavate how automation and AI changes the jobs (and, yes, AI can be used to scale this process) This enables a strategy for L&D to be in service of the most critical aspects of business continuity. For the first time in L&D's history, we face the daunting task of simultaneously preparing the workforce to execute strategies resulting from automation and AI while preventing the instability that a skills flux brings to the business and the workforce. Here are links to these two reports: -- Indeed Hiring Lab: https://lnkd.in/grF2C2-E -- Kyla Scanlon: https://lnkd.in/gAkcj4Qi

The GenAI Layoff Myth: They Are Not Telling You The Truth Andi Mazingo "There's a sucker born every minute," often attributed to P.T. Barnum, underscores the ease with which we can be misled. This sentiment resonates as companies attribute layoffs to the rise of Generative AI (GenAI), diverting attention from managerial shortcomings. In times of economic uncertainty or declining profits, companies often resort to cost-cutting measures, with layoffs being a primary tool. Traditionally, such decisions stem from internal factors: mismanagement, poor strategic planning, or failure to adapt to market changes. By attributing job cuts to technological advancements, companies can: Shift Responsibility: Blaming AI diverts scrutiny from leadership decisions. Appease Investors: Positioning layoffs as forward-thinking can bolster investor confidence. Mitigate Backlash: Technological inevitability is harder to contest than managerial incompetence. Contrary to the alarming headlines, empirical studies suggest that GenAI's impact on employment is more nuanced. 1. The Chicago Booth Study A study from the University of Chicago's Booth School of Business found that AI chatbots, a subset of GenAI, have had no significant impact on earnings or recorded hours in any occupation. The researchers concluded: This finding challenges the narrative that GenAI is a primary driver of job displacement. 2. Forrester's Forecast Forrester's 2023 study projected that only 1.5% of jobs in the U.S. would be lost to GenAI by 2030, while 6.9% would be influenced by the technology. The decision to lay off employees under the guise of technological advancement often masks deeper issues within an organization. These may include: Strategic Missteps: Entering unsustainable markets or failing to innovate. Inefficient Operations: Poor resource allocation or outdated processes. Leadership Failures: Inability to inspire, lead, or make informed decisions. By attributing layoffs to GenAI, management can deflect criticism and avoid addressing these core issues. This tactic not only undermines employee morale but also erodes trust in leadership. The portrayal of GenAI as a primary catalyst for job losses lacks substantial empirical support. Instead, this narrative often serves as a smokescreen for managerial shortcomings and strategic failures. It's imperative for stakeholders, employees, investors, and society at large, to critically assess such claims and hold leadership accountable. ******************************************************************************** The trick with technology is to avoid spreading darkness at the speed of light Stephen Klein is Founder & CEO of Curiouser.AI, the only Generative AI platform and advisory focused on augmenting human intelligence through strategic coaching, reflection, and values-based decision-making. He also teaches AI Ethics at UC Berkeley. Learn more at curiouser.ai or connect via Hubble https://lnkd.in/gphSPv_e