When a CS professor at IE University starts his argument with "trust me, I'm a computer scientist," you know you're about to hear the academic equivalent of a Blockbuster executive explaining why Netflix is just a fad. Professor Ikhlaq Sidhu's rosy manifesto reads like it was written by someone who hasn't looked at a hiring board in 2025, hasn't talked to a junior developer in months, and definitely hasn't watched Mark Zuckerberg casually announce that AI will be doing mid-level engineering work this year.
We get it, Professor. Nobody wants to be the messenger of their own profession's obsolescence. But denial isn't just a river in Egypt—it's apparently the primary water source for computer science departments worldwide.
The Numbers Don't Lie, But Academics Apparently Do
Let's start with the most glaring oversight in Sidhu's argument: he completely ignores the employment data that's been screaming at us for months. Big Tech companies reduced new graduate hiring by 25% in 2024 compared to 2023, while IT sector unemployment jumped from 3.9% to 5.7% in just one month—far above the national average. 152,000 IT jobs were lost in January 2025 alone, and these aren't just pandemic-related cuts. These are structural changes.
At Google, over 41% of all code is now AI-generated, and Salesforce CEO Marc Benioff announced the company would pause hiring software engineers in 2025 due to a "30% productivity boost" from AI tools. Meanwhile, Meta CEO Mark Zuckerberg stated that AI will perform mid-level engineering tasks in 2025. These aren't theoretical predictions—they're quarterly earnings announcements.
But sure, Professor, tell us more about how AI "cannot reliably" do complex systems work while CEOs are literally replacing their engineering teams with algorithms.
Sidhu's most catastrophic blind spot is his complete failure to address what's happening to junior developers. A survey of 9,000 software engineers found that 90% believe finding a job is significantly harder than it was in 2020, with only 6% confident they could match their current salary if they lost their job. Many businesses find that AI-assisted coding allows mid-level and senior engineers to be more productive, reducing the need to hire junior developers.
Here's the uncomfortable truth that CS professors refuse to acknowledge: one experienced engineer working alongside AI can now do the work of what used to be a three-person team. 25% of developers estimate that 1 in 5 AI-generated suggestions contain factual errors, but companies are willing to accept that error rate in exchange for massive cost savings.
The pipeline is broken. Junior developers historically learned by doing repetitive tasks, debugging, and writing boilerplate code—exactly the work AI now handles. With AI generating code for standard functions and simple applications, automating quality testing and debugging, junior staff are more affected than seniors. Without entry-level positions, where exactly will the next generation of "senior" developers come from? Spontaneous generation?
Here's where Sidhu's argument becomes almost comically outdated. Despite 2024's heavy investment in AI-generated code, developers still aren't seeing marginal gains—speed and stability have actually decreased due to AI, according to the 2024 DORA report. But companies don't care about code quality when they can cut payroll by 30%.
Developers who use AI tools for coding are 88% more productive than those who don't, but "productivity" in Silicon Valley means "fewer humans needed." 76% of developers are using or planning to use AI tools, up from 70% the previous year, but favorability declined from 77% to 72%—possibly due to disappointing results from usage.
The writing is on the wall in binary: AI doesn't need to be perfect to replace jobs. It just needs to be cheaper.
Sidhu trots out the tired industrial revolution analogy, suggesting that technical skills will be "more valuable than ever." This is exactly backward. During the Industrial Revolution, factory workers were displaced at a 50 to 1 ratio, but the workforce grew because most new workers operated or fixed machines. The key difference? Those industrial workers weren't teaching machines to build better machines.
AI could eliminate half of all entry-level white-collar jobs within five years, according to Anthropic CEO Dario Amodei. 77% of AI jobs require master's degrees, and 18% require doctoral degrees. We're not creating a broader technical workforce—we're creating an extremely narrow elite while everyone else gets automated out.
Sidhu's most intellectually dishonest argument is his claim that AI "cannot reason, feel, care, or desire anything." Neither can most entry-level programming tasks. Developers spend only 24% of their time writing code, according to Forrester's 2024 survey, but that 24% is what pays the bills. The other 76%—system design, stakeholder collaboration, creative problem-solving—requires experience you can't get if AI eliminates all the entry-level positions.
Research from Anthropic analyzing a million conversations with Claude found that people used AI to perform programming tasks more than any other occupation. The "AI can't think" argument crumbles when companies are already using AI to think well enough to replace human workers.
Professor Sidhu's argument essentially boils down to: "AI can't do the hard stuff, so CS is safe." But he's fighting the wrong battle. AI doesn't need to solve nuclear reactor software or quantum computing—it just needs to eliminate enough routine work to make most CS positions economically obsolete.
41% of employers worldwide intend to reduce their workforce due to AI automation, and up to 30% of jobs could be automatable by the mid-2030s. The question isn't whether AI will eventually match human creativity and reasoning. The question is whether there will be enough economically viable CS jobs left by the time we find out.
Academia loves to focus on the theoretical while ignoring the practical. But when Zuckerberg says AI will do mid-level engineering work in 2025, and Benioff stops hiring engineers entirely, maybe it's time for CS professors to spend less time explaining why AI can't work and more time explaining why their students can't find jobs.
The market has spoken. It's choosing cheaper AI over expensive humans, flawed algorithms over perfect code, and short-term savings over long-term talent development. Professor Sidhu can keep telling his students that computer science is indispensable, but the companies writing the paychecks have already made their decision.
Need help positioning your marketing team for an AI-driven economy where technical skills are increasingly automated? Winsome Marketing's growth experts understand how to build resilient marketing strategies that leverage AI rather than compete with it. Get in touch.