David Linthicum

This video breaks down a new paper, Remote Labor Index: Measuring AI Automation of Remote Work, and why it challenges the biggest claims in the AI hype cycle. Instead of scoring AI on narrow tests, the researchers evaluated frontier systems on 240 real freelance projects across 23 categories of paid remote work, using actual briefs, input files, and human-created deliverables as the benchmark. The result was brutal: current AI agents performed near the floor, with the best system reaching only a 2.5% automation rate.

In other words, even the strongest model could complete only a tiny fraction of projects at a level a reasonable client would accept as commissioned work. The study also found widespread failure modes, including poor quality, incomplete outputs, corrupted files, and inconsistent deliverables. That matters because these were not toy tasks; the benchmark represented over 6,000 hours of real work and more than USD 140,000 in project value. The takeaway is not that AI has zero value, but that the real-world automation story is far weaker than the hype suggests, especially when judged on end-to-end professional freelance work. It shows progress is real, but broad autonomous labor replacement is still much more promise than present-day reality today.

In this video, I dive into McKinsey's headline-grabbing claim of employing 25,000 AI "agents" as if these bits of software are somehow interchangeable with human employees. Let's be blunt: this reveals how little McKinsey—and much of big consulting—actually understands about what AI agents are, what they aren't, and where their real value lies. What we're seeing is not a revolution in talent strategy, but a parade of buzzwords masking a fundamental confusion. AI agents aren't employees. They don't manage liability, contribute to culture, develop critical thinking, or own outcomes. Treating automation like headcount distorts decision-making, creates risk, and buries the real discussion about where AI can drive measurable impact.

McKinsey is pitching a future where platforms and agent orchestration are the answers to everything—when, in reality, clients get locked into over-engineered, expensive stacks that rarely deliver. True innovation starts with a business-first, problem-first approach: map processes, clarify goals, and select the right tool for the job—AI or otherwise. In this video, I break down the absurdity of pretending AI agents are employees, explain why this mindset is holding consulting back, and lay out how companies should really be thinking about the intersection of automation, analytics, and human expertise.

AI agents are reshaping how people interact with business software—and that's why you're hearing claims like "AI will replace SaaS." The core idea is simple: instead of employees logging into five different tools, learning each interface, and clicking through workflows, they'll describe the outcome they want and an agent will execute the steps across systems. When that happens, the seat-based revenue model weakens because fewer humans need direct app access. At the same time, many point-solution products risk becoming interchangeable "capabilities" that the agent can reproduce or route around, pushing buyers to consolidate spend into the platform layer that owns the agent interface and distribution. In this new setup, some SaaS tools survive as trusted backends—systems of record with strong data, governance, permissions, and compliance—while others get squeezed into lower-margin API roles. The likely reality isn't that software disappears; it's that the UI layer changes, packaging shifts from seats to usage or outcomes, and competitive advantage moves toward proprietary data, deep workflows, and distribution. This video breaks down the before-and-after model, why the debate is happening now, and what it means for SaaS pricing, churn, and category winners.

Block just announced another round of layoffs, and the loudest part of the story isn't the headcount number—it's the narrative. The company is leaning into an "AI-driven efficiency" framing, implying that automation is the primary force reshaping teams and roles. But there's a much simpler (and more honest) explanation: overhiring followed by cost normalization.

In this video, we break down why "AI made us do it" is quickly becoming the most convenient corporate cover story of 2026. AI can absolutely change workflows, reduce manual work, and shift what kinds of roles are needed—but it doesn't magically create a bloated org chart. Hiring ahead of demand does. And when growth doesn't match expectations, the correction shows up as layoffs, reorganizations, and "efficiency" initiatives.

We'll connect the dots between post-boom expansion, management layers, duplicated scope, and the push to protect margins—then explain why blaming AI is both strategically useful and culturally dangerous. If we don't separate tech reality from leadership decisions, we'll misunderstand what's happening across the entire job market.

AI is everywhere, but the loudest takes aren't always the most useful. In this video, we break down five unpopular (and highly practical) opinions on AI drawn from David Linthicum's enterprise-focused writing—covering why "agentic AI" may be running ahead of real-world execution, why chasing hype can quietly drain budgets, and why simpler AI often delivers better near-term ROI than flashy initiatives. We also tackle the uncomfortable truth about AI returns: many organizations struggle to measure value, lack the data foundations to scale, or underestimate the operational lift required to make AI stick. Finally, we look at why cloud-based AI can feel far harder than promised—where data governance, skills gaps, and integration complexity become the real blockers.

If you're a leader, builder, or just trying to separate signal from noise, this is a grounded reality check with clear takeaways you can apply to strategy, tooling, and prioritization. Watch to the end for a simple checklist to sanity-test your AI roadmap before the next big spend. Like, subscribe, and comment: which opinion will age best?

Anthropic just raised the alarm: it says a network tied to Chinese AI companies created thousands of accounts to hammer Claude with prompts, capture the outputs, and use them to train competing models—a technique known as model distillation. If true, this is a new kind of "theft": not a server hack, but a high-volume scraping of a model's capabilities through the front door of an API. In this video, we break down what distillation is, why it threatens frontier-model IP, and how it can quietly shift the balance in the global AI race. We'll cover the alleged tactics (fake accounts, automation, evading rate limits), the defenses (behavioral detection, throttling, watermarking, and policy changes), and the bigger stakes for startups, national security, and open research. Most importantly, we'll ask: how do you protect an AI model when the product is an interface and the "asset" is knowledge embedded in weights? Watch to the end for practical safeguards any team shipping an LLM can implement today—and what this episode signals about what comes next. We'll also separate hype from evidence, explain what companies can log without violating privacy, and discuss whether regulation, contracts, or technical friction will matter most in the future.

For the last few years, Big Tech pushed a blunt message: AI will replace developers, automate entire departments, and become systemic to every part of life. That story drove panic, boardroom decisions, and layoffs framed as "efficiency" or "AI transformation." But here we are in 2026, and the promised overhaul still hasn't arrived.

In this video, I break down what was claimed, why it was claimed, and what actually happened. You'll see how the AI narrative split into two groups: the insiders who understand the limitations but keep selling the dream because revenue and valuation depend on it, and the downstream leaders who repeat the messaging because their businesses now rely on AI being "the future."

This isn't an anti-AI rant. It's a reality check. We'll talk about what AI is genuinely good for today (narrow, high-leverage assistance), where it fails (end-to-end replacement of complex engineering work), and how to protect yourself and your organization from the next hype cycle.

Amazon just confirmed plans to cut roughly 16,000 roles, including in AWS—another reminder that even "safe" high-tech jobs can vanish fast. If you're being pushed out (or fear you might be), this video gives you a clear, practical playbook to regain control.

First: a layoff is a business event, not a personal verdict. Your job is to stabilize your mindset and your runway so you can make smart decisions quickly. Next: stop "job hunting" and start value positioning. The market doesn't pay for buzzwords; it pays for outcomes—cost removed, risk reduced, reliability improved, delivery speed increased.

Then choose one lane and go deep. Panic-skill-spamming with random certs wastes time. Focus on work that survives budget pressure: FinOps and cloud cost optimization, security/governance/compliance, platform engineering and reliability, modernization with real ROI, and production-grade data engineering.

Finally, treat your search like operational excellence: a 30–60–90 day plan, shipped artifacts, targeted outreach, and interview stories tied to measurable impact. If you're navigating a layoff right now, you're not alone—and you're not powerless.

Subscribe for balanced AI + cloud takes.

The U.S. Federal Trade Commission (FTC) is intensifying its investigation into Microsoft's dominance in the world of business software, cloud computing, and artificial intelligence. With a sharp focus on Microsoft's widely used products like Windows, Office, and its AI Copilot, regulators are probing claims that Microsoft may be leveraging its market power to unfairly push AI offerings onto its users and lock out competitors. Subpoenas have gone out to at least six rival companies, demanding deep insight into Microsoft's licensing, bundling, and cross-platform compatibility practices.

Central to the probe is whether Microsoft's integration of AI, security, and identity services into its dominant platforms effectively stymies fair competition by making it harder for customers to use competing cloud services or alternative software. Launched under Lina Khan during the Biden Administration and continuing under the Trump Administration's FTC Chairman Andrew Ferguson, this sweeping inquiry marks one of the toughest regulatory challenges to Microsoft's grip on enterprise technology since the 1990s. As the FTC consults with various companies and industry experts, the growing scrutiny signals that regulators are prepared to confront tech giants who may be using their position to reshape user choice and the future of AI and cloud technology.

AI regulation is moving fast—and it's not happening the same way everywhere. In this video, we break down what's underway right now across the EU, UK, US (federal + states), and Canada in a clear, non-hype way. You'll learn how the EU AI Act sets a risk-based framework (from banned practices to strict controls for high-risk systems and new rules for foundation models), while the UK leans on existing regulators and principles instead of one mega "AI law." In the United States, there's still no single national AI statute, so the real action is in agency guidance, enforcement, and government procurement rules—plus a rapidly growing state-by-state patchwork. We'll also zoom into major state momentum, including California's combination of privacy governance and AI transparency/synthetic media rules, and what's developing in Florida through targeted privacy and anti-deception approaches.

Finally, we cover Canada's direction with proposed "high-impact AI" requirements (AIDA) and how privacy law already shapes AI deployment. Whether you build AI products, buy AI tools, or just want to understand where policy is headed, this is your quick map of the emerging rules of the road. Subscribe for updates as these laws evolve.