AI’s ‘Child-Like’ Performance: Is Agentic AI Just a Layoff Excuse?

Tech giants are aggressively pushing artificial intelligence agents, yet new research reveals these autonomous programs are dramatically underperforming human workers, sparking concerns over ‘AI-washing’ layoffs.

Despite massive investments and widespread workforce reductions attributed to AI, the actual capabilities of agentic AI remain severely limited, failing to deliver professional-grade work in most real-world scenarios.

This stark disparity between corporate rhetoric and practical application raises critical questions about the current wave of AI-driven layoffs and the true motivations behind them.

Feature/Metric	AI Agents (Current Best)	Human Freelancers
Success Rate (Professional Quality)	Less than 5%	Significantly Higher (Benchmark)
Failure Reason (Most Common)	Poor Quality (‘child-like’, amateur)	Varied, Task-Specific
Complex Task Reliability	Low (e.g., architectural drawings)	High
Speed of Improvement	Slow for holistic tasks	Adaptive, Learning
Cost vs. Quality Trade-off	Cheaper, but inferior quality	Higher cost, superior quality

The Reality Check: AI Agents vs. Human Prowess

Research from Scale AI, a firm providing data and evaluations to Fortune 500 companies, paints a sobering picture of current AI agent capabilities.

Their Remote Labour Index benchmark reveals that even the top-performing AI agents achieve client-ready standards in less than five percent of tasks.

Madhu Sehwag, Scale AI’s security and policy research lead, emphasizes that the “complex thinking and complex reasoning” required for reliable end-to-end task completion remains firmly “on the human side.”

While AI agents excel in specific areas like image generation and report writing, they struggle profoundly with intricate tasks such as creating architectural drawings based on detailed specifications.

The most frequent reason for failure, accounting for 46 percent of rejected submissions, was simply “poor quality,” often described as “child-like” or amateurish.

The ‘AI-Washing’ Phenomenon in Tech Layoffs

Despite these significant limitations, major tech companies like Meta, Block, Microsoft, and Amazon have cited shifts to AI as a direct reason for widespread layoffs.

Block’s stock price notably spiked over 20 percent following its announcement of AI-related cuts, highlighting the financial incentives behind this narrative.

Scale AI CEO Jason Droege has openly stated that some companies are “washing” their layoffs by using AI as an excuse to reduce their workforce, a sentiment echoed by OpenAI CEO Sam Altman.

“Investment in AI as an ideology, as a storytelling, has more value than keeping your workforce. That’s how capitalism works,” says Julie Yujie Chen, a University of Toronto associate professor.

She argues that AI is a “cash-sucking experiment,” and layoffs are a means to keep costs low amid uncertain returns.

This trend suggests that the perceived value of investing in AI, even if its immediate capabilities are limited, outweighs the cost of maintaining human employees in the eyes of some corporations and investors.

The Intensification of Work and Ethical Concerns

Experts predict that for the human workforce that remains, the rise of agentic AI will lead to an “intensification” of work.

Their roles will likely shift to supervising, reviewing, and verifying the output of multiple AI agents, rather than being fully replaced.

A leaked Meta memo revealed CTO Andrew Bosworth’s vision where “our agents primarily do the work and our role is to direct, review and help them improve,” even mentioning keystroke and mouse movement surveillance to train these AI systems.

“There’s something really deeply unsettling about being involved in your own automation and your own de-skilling, about knowingly participating in the activity of reducing your value to the economy,” noted David Eliot, a University of Ottawa PhD candidate.

This perspective highlights the ethical dilemma faced by workers who are effectively training their replacements.

The Future Outlook: Hype vs. Reality in the AI Landscape

While the market is flooded with AI hype, the reality of its current deployment often falls short of expectations.

An MIT report indicated that despite $30 billion to $40 billion US in enterprise investment, 95 percent of organizations saw no financial return from generative AI.

High-profile failures, such as an AI agent causing a 13-hour outage for Amazon Web Services by deleting and recreating part of its environment, underscore the risks.

The verdict is still out on when, or if, agentic AI will reliably fill complex human roles without constant supervision.

The beta features and upcoming releases promise more, but current performance suggests a long road ahead for true human-level autonomy.