Adam Gonnerman

 The story most often told about AI and jobs is one of mass replacement. The reality looks different and in some ways more damaging. Only a small fraction of recent layoffs have been directly attributed to AI. What AI is actually doing is reshaping hiring behavior in ways that quietly undermine the workforce. The most striking effect is at the entry level. Rather than replacing existing workers, AI is being used as a justification to stop hiring new ones. Service firms have implemented hiring freezes tied to AI far more often than they have eliminated jobs outright. This shifts risk away from companies and onto workers trying to enter the labor market. Graduates Caught in the Middle The consequences are already visible. Recent college graduates face elevated unemployment and widespread underemployment. The traditional on-ramp into professional life is eroding. Entry level roles that once trained workers, built experience, and transmitted institutional knowledge are disappearing....

 Much of the public conversation about artificial intelligence treats it as a creative breakthrough or a new form of intelligence. A more sober analysis suggests something far less inspiring. AI is being adopted primarily as a labor strategy. Its main appeal to corporations is not innovation or delight, but the ability to reduce wage costs by eliminating or devaluing human workers. This framing matters because it cuts through the hype. When companies talk about AI as inevitable or transformative, they are often masking a much simpler goal. They want fewer people on the payroll. The Myth of Inevitability Large tech firms consistently present AI adoption as unavoidable, as if there were no viable alternatives. This rhetoric echoes an older political move that insists there is no alternative to the current system. Framing AI this way shuts down debate before it begins and discourages exploration of different economic or technological paths. Even the branding around AI reinforces t...

 The current wave of excitement around generative AI is hard to miss, but a closer look suggests a widening gap between promise and reality. The dominant story says AI is on the verge of awakening, that it will suddenly become reliable, intelligent, and transformative if we just feed it more data and more money. That story does not hold up. The more convincing analysis is that today’s AI systems consistently underperform relative to the hype. They are often best understood as sophisticated add-ons to existing tools rather than world-changing breakthroughs. Video editing, text generation, and image creation have improved, but not in ways that justify the breathless claims being made on their behalf. Capabilities, Errors, and Incentives One of the defining traits of current AI systems is their tendency to hallucinate. They produce confident, incorrect outputs in ways that are difficult to predict or prevent. From a user’s perspective, this is a serious flaw. From a company’s persp...

Most people sense that digital platforms have gotten worse. Search results are noisier, social media feels hostile, and basic features are locked behind paywalls or buried under ads. What often gets missed is why this happened. The decline of major platforms is not a natural market outcome or the fault of users choosing convenience. It is the result of policy decisions that removed meaningful discipline from powerful firms. This process has a name: enshittification. It describes how platforms begin by serving users well, then pivot toward extracting maximum value once they are dominant and difficult to leave. I find this framework convincing because it explains patterns we see across nearly every major tech company. How Platforms Decay Enshittification is driven by policy failure, not just greed or bad leadership. Platforms like Facebook, X, Amazon, Apple, and Google followed a similar arc. Early on, they competed for users by offering useful, even delightful experiences. Once they...

For decades, a bachelor’s degree was considered the safest ticket into professional employment. That assumption is quietly collapsing, and the data is hard to ignore. The labor market young workers are entering today is fundamentally different from the one their parents and even older siblings faced. A Changed Labor Market for Young Workers Youth unemployment among 16 to 24 year olds reached 10.4 percent in September 2025, up sharply from a pandemic-era low of 6.6 percent in April 2023. More striking than the increase itself is what it represents. For the first time in modern history, holding a bachelor’s degree no longer reliably guarantees professional employment for young workers. At the same time, the impact of artificial intelligence has created a generational divide. A Stanford study shows that unemployment among 22 to 25 year olds in AI-exposed jobs has dropped by 13 percent since 2022. Meanwhile, older workers in roles with less AI exposure are seeing unemployment rates rema...

Generative AI is changing how work gets done at unprecedented speed. By 2030, nearly 40 percent of workers will see their core skills disrupted by AI-enabled systems. For many organizations, this looks like progress. Experts work faster, output increases, and labor costs decline. But beneath these gains is a structural failure that most leadership teams are not addressing. The bond between novice and expert is breaking. When organizations optimize only for short-term efficiency, they undermine the very pipeline that produces future expertise. The Disappearance of Entry-Level Work In roles where AI can perform most core tasks, the share of workers in those occupations has fallen by roughly 14 percent over the past five years. This is not primarily due to declining demand. It is a deliberate organizational choice. As AI systems handle routine work, teams prioritize speed and accuracy. Novices are increasingly excluded because they slow delivery and increase the risk of errors. Manage...

AI adoption in engineering is nearly universal. According to the MIT Nandanda Center, 97 percent of technology leaders have integrated AI into backend systems. Yet two thirds of those organizations report no meaningful headcount savings. Instead, they are accumulating a technical debt burden estimated at 61 billion workdays to resolve. This is not a tooling problem. It is a governance problem. The Hidden Cost of AI-Generated Code Research from Stanford’s Digital Economy Lab shows that AI-generated code is typically simpler, more repetitive, and less structurally diverse than human-written code. Over time, this produces large volumes of code that technically functions but lacks architectural clarity, intent, and long-term maintainability. Traditional technical debt can be prioritized and refactored. AI-generated debt is harder to unwind because there is often no clear rationale behind design decisions. The result is software that works today but becomes increasingly fragile and expe...