Home/News/AI Agents Era: Companies Hired 'Bad Employees' Without Management
Fortune3 min read

By Interestana AI Editorial — AI-drafted, human-overseen. How we report

AI Agents Era: Companies Hired 'Bad Employees' Without Management

AI Agents Era: Companies Hired 'Bad Employees' Without Management

George Sivulka, founder of AI enterprise startup Hebbia, stated that companies have rushed to deploy AI agents without establishing the necessary management infrastructure. Hebbia, which serves clients like BlackRock, KKR, and the U.S. Air Force, has provided Sivulka with direct insight into the challenges enterprises face with AI adoption. In an essay published on a16z's newsletter, Sivulka contends that AI has not reduced labor costs but rather inverted the cost dynamic, making humans cheaper than software for the first time. He warned CEOs who embraced the "tokenmaxxing" era that they effectively hired "a million bad employees."

The "tokenmaxxing" era appears to have concluded following Amazon's disclosure of a $500 million monthly loss attributed to unmanaged AI agents. Concurrently, Ford Motor Company has reportedly rehired human engineers, referred to as "graybeards," to collaborate with AI augmentation efforts. Sivulka draws a parallel between this period and a significant railroad crash in 1841, which he posits marked the end of one era and the beginning of another.

Sivulka's historical analogy points to the 1830s and 1840s, a period of rapid expansion in American railroad mileage. The lack of coordinated systems led to a fatal train collision in Massachusetts in 1841, which in turn spurred the development of modern management practices, including defined roles, reporting lines, and hierarchies. This crisis was instrumental in transforming railroads into a major industry, accounting for approximately 60% of the stock market at its peak.

He argues that just as railroads facilitated widespread travel, AI agents have similarly transformed how work is conducted online. Sivulka explained that companies have granted every employee, regardless of performance, effectively unlimited headcount and budget. He emphasizes that managing AI is more complex than managing people because AI can instantly scale dysfunction. This rapid scaling of inefficiencies, without proper oversight, has become a significant challenge for businesses.

Original source — read the full reporting at the publisher:

Read on Fortune

Get the weekly AI digest

AI news + new model releases, weekly. Drafted by our agents, reviewed by humans.

Read next