The Moral Mess of AI: How Tech’s Automation Obsession is Risking Jobs and Innovation

  

Fudzilla (2024). US IT jobs slump as AI takes over.

 

Artificial intelligence (AI) has become deeply embedded in daily life, from ChatGPT assisting with complex mathematics to Netflix recommendations shaping entertainment choices. While AI enhances efficiency, its rapid rise is often driven by cost-cutting strategies, raising concerns about job security in an increasingly automated world. This shift is evident in tech giants such as Meta – the powerhouse behind Facebook, Instagram and WhatsApp, which is aggressively hiring AI specialists while laying off thousands of employees. This time, 3,600 workers - 5% of its global workforce, branded as ‘low performers.’ This follows over 21,000 job cuts in 2022 and 2023 (BBC, 2025). But here's the big question: what if those so-called underperformers are actually critical problem solvers? By putting automation ahead of human expertise, Meta risks sacrificing long-term innovation for short-term efficiency. As artificial intelligence becomes the focal point of technology strategy, the workforce is being changed, prompting concerns about fairness, responsibility, and the true cost of efficiency.

Moral Hazard: Hidden Risk in a High-Tech Gamble

Why is Meta doing this? Meta's choice to reduce jobs while aggressively employing AI professionals is more than just a cost-cutting measure; it's a textbook example of moral hazard. Moral hazard occurs when decision-makers take significant risks without considering the consequences. In this situation, Meta's leaders are protected from the consequences of their restructuring actions. If AI bets fail, thousands of people will be laid off, not CEOs. And Meta is not alone. Leaders across the tech industry are shifting their focus to automation, chasing quarterly victories, and betting on algorithms (Clark, E. 2025). However, without accountability, these leaders may overinvest in trends while undervaluing human capital, undermining the trust and stability required for long-term innovation. The trend parallels earlier bubbles, such as the dot-com catastrophe and subprime mortgages, in which short-term risk-taking resulted in long-term consequences. If history tells us anything, it's that unrestrained incentives breed systemic fragility.

 

 

 

The Fog of Evaluation: Asymmetric information and Bias.

This risk-taking under uncertainty is further muddled with asymmetric information—wherein employees typically have more information on their true contributions than the managers evaluating them. The logic behind Meta's firings—labeling employees as "low performers"—relies heavily on broken performance management systems. Many employees reported having been ambushed by unexpected demotions, despite having good track records of contribution. These systems often cannot measure creativity, collaboration, or enduring problem-solving. And, rather than improving clarity, incorrect measures exacerbate moral hazard. Leaders, misled by insufficient data, make broad judgments with little comprehension of who they're removing—or what those people actually brought to the table. Worse, individuals who excel at office politics are frequently retained, while behind-the-scenes contributors are dismissed. That not only lowers morale, but also creates inefficiency, which Meta claims to be eradicating (Clark, E. 2025).

Layoffs by the Numbers: Optics vs. Reality

This moral hazard is compounded by another issue: how Meta decides who remains and who leaves." Productivity scores, peer assessments, and manager feedback—all of which are imperfect. They frequently prioritize exposure over impact, preferring loud staff over quiet top performers. Bias comes in, and valuable contributors are mistakenly classed as expendable. This error not only reflects bad data, but it also contributes to moral hazard. Leaders make high-stakes decisions based on flawed systems without realizing the repercussions. Meanwhile, the workers who are most important to innovation—and have the least ability to protect themselves—are cut. Shouldn't we be questioning if your analytics can detect creativity or collaboration? Are you truly evaluating performance? Or are you simply optimizing for optics?

Efficiency’s Hidden Price Tag: The Numbers Behind the Shift

 

Macrotrends (2024). Meta Platforms Financial Statements

 

Despite the massive layoffs, Meta's profits are increasing. Why? Machines do not beg for sick days. AI does not require benefits. With automation in place, expenses fall while output remains constant—or even increases. Meta's operating expenses decreased from $62.4 billion in 2022 to $41.1 billion after lowering headcount by more than 24,000. That's a staggering $21.3 billion in savings. Meanwhile, Meta's revenue climbed from $116.6 billion in 2022 to $134.9 billion in 2023, a 15.7% increase, while net income increased from $23.2 billion to $39.1 billion, up 68.5%. These figures highlight the financial motivation for automation, but they also obscure the true human costs. Who pays the price if executives are never affected by a bad AI gamble? It is more than just a technological update; it is a risk transfer from the powerful to the disadvantaged. And when leaders are rewarded for short-term gains, what incentive do they have to think about long-term consequences?

Who Gets Left Behind: Low-Income Workers at Risk

 

UK Government (2021) Page 8

 

Meta may be leading the charge, but it is far from alone. Across industries, a similar pattern is emerging: AI threatens not only jobs, but also the workers who are least prepared to adapt. Routine and repetitive work are being automated, including sales assistants, call center personnel, and even cleaners. According to the data, these roles are most likely to disappear within the next two decades. Jobs that need creativity and sensitivity such as nurses, engineers, and CEOs, are likely to rise. The lesser the income, the greater the likelihood of automation. This puts low-income workers in the crosshairs of a revolution they are ill-equipped to withstand. Without training, regulation, or safety nets, AI risks exacerbating inequality. A future of efficiency should not leave whole communities behind. Up to 30% of current employment are at risk of automation by the mid-2030s. That includes 44% of workers with low education, compared to just 14% of those with higher education (UK Government, 2021)

AI is revolutionizing sectors, yet businesses such as Meta favor automation above human expertise, jeopardizing long-term innovation and exacerbating inequality. Flawed evaluation methods misclassify vital workers as replaceable, shifting risks to employees while executives gain the benefits. Without accountability and equitable labor rules, artificial intelligence may replace, rather than complement, human intellect. The future of work depends on responsible integration, not just efficiency.

References

Fudzilla (2024). US IT jobs slump as AI takes over. https://www.fudzilla.com/news/ai/58223-us-it-jobs-slump-as-ai-takes-over

BBC News (2025). Meta job cuts: Roughly 3,600 people could be affected.  https://www.bbc.co.uk/news/articles/c3e18lnl20po#:~:text=Roughly%203%2C600%20people%20could%20be,about%2011%2C000%20roles%20in%202022

Clark, E. (2025, February 11). Meta's Job Cuts Add To Its Ruthless Culture. Startups.co.uk.  https://startups.co.uk/news/meta-job-cuts-company-culture/

Macrotrends (2024). Meta Platforms Financial Statements. Macrotrends. https://www.macrotrends.net/stocks/charts/META/meta-platforms/financial-statements

UK Government (2021). The impact of AI on jobs. GOV.UK.  https://assets.publishing.service.gov.uk/media/615d9a1ad3bf7f55fa92694a/impact-of-ai-on-jobs.pdf