Labor Market Impact of AI Chatbots

Call it Service-as-a-Software. Call it AI agents or agentic systems. There’s a brewing idea that AI will complete human labor especially in white-collar work. What attributes of a market make it attractive to pursue? Those with three attributes : Toil, labor market shortages, and margin pressure. Toil is repetitive work : reviewing alerts, triaging leads, data entry. Necessary but not strategic. Jobs laden with toil tend to be difficult to recruit for & retain. Turnover rates of 30-50% are common in these roles. Labor market shortages are a result of a mismatch between the supply of labor and the demand for it. Perhaps not enough graduates in a particular discipline. For example, accounting graduates have fallen approximately 18% since 2016. or too few applicants for a particular role like customer support. Whatever the reason, the challenge is the same facing a hiring manager : difficult recruitment to maintain or grow headcount. Last, margin pressure. Wobbles in the economy are impacting the labor market. Unemployment is now at 4.3% & new job creation has fallen in half compared to the last 12 months employers will need to do more with less. Recent earnings reports from publicly traded companies that use AI continue to underscore the significant cost savings when AI is deployed successfully. Last week, Amazon reported the impact of its AI system called Q : “With Q’s code transformation capabilities, Amazon has migrated over 30,000 Java JDK applications in a few months, saving the company $260 million and 4,500 developer years, compared to what it would have otherwise cost.” ServiceNow mentioned British Telecom (BT) : “BT Group announced that its now-assist pilot helps agents write case summaries and review complex notes faster, cutting both times by 55 percent. This helped drive down the average time to resolve cases by one-third.” The ideal customer profile for an AI startup are hiring managers recruiting for rote work in challenging labor markets facing margin pressure. When faced with the choice between a long hiring process or the potential to fulfill the role with a software robot at 15-20% the cost of human labor, a hiring manager calculated risk to try AI may result in tremendous savings to the business.

Worrying: "We find that since the widespread adoption of generative AI, early-career workers (ages 22-25) in the most AI-exposed occupations have experienced a 13 percent relative decline in employment even after controlling for firm-level shocks. In contrast, employment for workers in less exposed fields and more experienced workers in the same occupations has remained stable or continued to grow... Our results are robust to alternative explanations, such as excluding technology-related firms and excluding occupations amenable to remote work. These six facts provide early, large-scale evidence consistent with the hypothesis that the AI revolution is beginning to have a significant and disproportionate impact on entry-level workers in the American labor market" By Erik Brynjolfsson, Bharat Chandar & Ruyu Chen at the Stanford Digital Economy Lab

Dario Amodei, CEO of Anthropic—the team behind Claude 4—says half of all entry-level jobs could vanish within five years. "We, as the producers of this technology, have a duty and an obligation to be honest about what is coming." That’s from Dario Amodei, CEO of Anthropic—the company behind Claude 4. He’s not talking about the distant future. He’s talking about the next one to five years. According to Amodei: > AI could drive U.S. unemployment to 10–20% > Half of all entry-level white-collar jobs could vanish—especially in law, tech, finance, and consulting >And "most of them are unaware that this is about to happen" He adds: "It sounds crazy, and people just don't believe it." Amodei is, appropriately, sounding an alarm. AI agents assist and replace. They are already being used to: >Code at human level >Review contracts >Analyze financial models >Handle customer support >Summarize medical records This story is already playing out: >Microsoft recently laid off 6,000 workers >Walmart is cutting 1,500 corporate roles >CrowdStrike cut 5% of its workforce, citing a tech inflection point "It's going to happen in a small amount of time—as little as a couple of years or less," Amodei says. Dario is calling for us to get prepared. He created the Anthropic Economic Index to track how AI is impacting jobs. He also launched an Economic Advisory Council to get ahead of the impact. And he’s floated a policy idea: a 3% "token tax" on model usage, redistributed to mitigate the fallout. He’s clear: "Obviously, that's not in my economic interest. But I think that would be a reasonable solution to the problem." The bigger concern is what happens to economic leverage—and democracy—when millions of people no longer generate value. "The balance of power of democracy is premised on the average person having leverage through creating economic value. If that's not present, I think things become kind of scary," said Amodei. What are you doing right now to prepare your team for what Amodei says is coming?

A new Stanford study has put hard data behind what many early-career professionals have been feeling: generative AI is disproportionately reducing entry-level job opportunities in fields like software engineering and customer support. The data is striking: 😢 Employment for workers aged 22–25 in the most AI-exposed roles has dropped by 13% since late 2022. 😄 Older workers in the same roles saw employment rise. ⭐ The biggest declines appear in jobs where AI is used to automate, not augment. ⭐ Salaries stayed flat — firms are cutting roles, not pay. This points to a deeper structural shift. AI appears to be replacing “codified” knowledge — the kind learned in school or bootcamps — faster than it can replace tacit, experience-driven skills. In other words: if your job can be learned from a textbook, it’s more replaceable. The result? The bottom rung of the career ladder is being sawed off. Without that first job, how does anyone gain the experience to climb? For leaders, this raises hard questions: ❓ How do we preserve pathways into high-skill careers? ❓Are we investing enough in human-AI complementarity, not just substitution? ❓What happens to organizations when new talent pipelines dry up? AI’s impact on work won’t be evenly distributed — and this may be one of the earliest, clearest fault lines. #AIWorkforce #EntryLevelJobs #FutureOfWork #AIEconomy #TalentPipeline #GenAI #Automation #AIImpact #LaborMarket #StanfordResearch

Generative AI Leads to Decline in Hiring for Exposed Jobs The introduction of ChatGPT in late 2022 sparked extensive research into the potential impact of generative AI on the labor market. With the increasing adoption of AI, there's an anticipation of significant workforce reductions in jobs highly susceptible to AI automation. Various researchers have attempted to identify which occupations are more vulnerable to AI: https://lnkd.in/gbmKT_Cz. It's important to note that workforce reductions could entail not just layoffs but also a decrease in hiring rates. Now, more than a year after ChatGPT's launch, we have sufficient data to begin examining whether there has been a reduction in hiring from 2022 to 2023 for jobs with high AI exposure compared to others. This is not a prediction, but an analysis of recent data. The main finding is that occupations with higher exposure to AI indeed experienced a greater decrease in online job postings between 2022 and 2023. However, the relationship between AI exposure and job posting declines is not straightforward. A notable reduction in job postings was only observed in the 40 percent of occupations most exposed to AI. In collaboration with Kimberly Kreiss, we conducted the following analysis: We performed a regression analysis across detailed occupations, with the percentage change in online job postings from 2022 to 2023 serving as the dependent variable. We controlled for industry variations to account for shifts in worker demand. The primary explanatory variable was the level of AI exposure. We used nine decile dummy variables to represent the levels of AI exposure, comparing their coefficients relative to the first decile. The outcomes of the decile coefficients are depicted in the chart below. #ai #generativeai #tech #recruitment #futureofwork

Employment for 22–25-year-olds in AI-exposed roles has dropped up to 20% since late 2022... A new Stanford report released today reveals that AI is already reshaping entry-level employment, and the first signs are in the data. The report, "Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of Artificial Intelligence," by Erik Brynjolfsson, Bharat Chandar, and Ruyu Chen, is the first large-scale empirical signal that AI is actively disrupting the labor market, and doing so unevenly. Analyzing ADP payroll data from 25 million+ U.S. workers, the report finds: ⭐ Employment for 22–25-year-olds in AI-exposed roles has dropped up to 20% since late 2022 ⭐ The shift isn’t limited to tech; trends are visible across industries and across data sets ⭐ Wages have remained stable, suggesting employers are cutting roles, not pay ⭐ The impact is concentrated in roles where AI automates, not where it augments That last point matters. Jobs that involve codified knowledge, like junior software development or customer service, are more vulnerable. Jobs that depend on tacit knowledge, collaboration, and judgment... less so. The researchers call young professionals in these roles the canaries in the coal mine. They’re not just early victims of automation, they’re early signals. So, if your organization is scaling AI, the strategic question isn’t just what we can automate. It’s whether we are building systems that replace talent or elevate it. The opportunity is still ours to shape. But only if we’re intentional. The report is robust, and I recommend downloading and reading it. It makes several additional important points. Download the report here: http://bit.ly/45Ttgzo

New employment data reveals a harsh reality for young professionals entering the workforce. Early-career workers aged 22-25 in AI-exposed occupations have experienced a 13% relative decline in employment since the widespread adoption of generative AI. The most affected roles? - Software developers - Customer service representatives - Clerical positions - Content creators - Business analysts What's concerning is that overall economic employment continues to grow. The jobs exist, but young people in these sectors aren't getting them. This isn't a temporary blip. Right now represents the worst AI will ever be. As the technology improves, we can expect this trend to expand beyond entry-level positions. But there's a silver lining for those willing to adapt. The same AI technology creating this displacement has also lowered the barriers to entrepreneurship more than ever before. Starting companies, building products, and creating services has become increasingly accessible. The question isn't whether AI will continue disrupting traditional employment paths. It's whether young professionals will compete or complain. ___________________ P.S. Follow me (Anthony Pompliano) for more insights on business, finance, & technology!