{"id":22481,"date":"2026-07-14T16:33:49","date_gmt":"2026-07-14T16:33:49","guid":{"rendered":"https:\/\/scannn.com\/you-just-hired-a-million-bad-employees\/"},"modified":"2026-07-20T15:35:38","modified_gmt":"2026-07-20T15:35:38","slug":"you-just-hired-a-million-bad-employees","status":"publish","type":"post","link":"https:\/\/scannn.com\/lv\/you-just-hired-a-million-bad-employees\/","title":{"rendered":"You just hired a million bad employees."},"content":{"rendered":"\n

AI was supposed to replace human labor.<\/span><\/p>\n

\n

It did the opposite.<\/p>\n

For the first time in history, humans are <\/strong>cheaper<\/strong><\/em> than software.<\/strong><\/p>\n

\n
\n
\"\"<\/picture><\/div>\n

 

Token spend per employee at top firms<\/figcaption><\/figure>\n<\/div>\n

And AI is creating <\/strong>more<\/strong><\/em> jobs than it eliminates.<\/strong><\/p>\n

\n
\n
\"\"<\/picture><\/div>\n

 

Headcount growth after AI adoption<\/figcaption><\/figure>\n<\/div>\n

In the 1830s, the advent of the railroad drove one of the largest infrastructure buildout the world had ever seen. American track mileage 120Xed in a decade.<\/p>\n

Then the system broke.<\/p>\n

On October 5, 1841, two trains fatally collided on the Western Railroad in Massachusetts due to a simple coordination failure.<\/p>\n

As the complexity of the railways grew, individual conductors were no longer enough to keep train travel safe. The railroad companies thus began a decades-long effort: hiring managers for each geography, defining roles in writing, and establishing clear hierarchies with reporting lines. Modern management was born. With it, rail became among the world\u2019s first billion-dollar industries, at its peak representing roughly 60% of the stock market.<\/p>\n

We just gave every employee, even the worst ones, effectively unlimited headcount and budget.<\/strong><\/p>\n

Managing AI is harder than managing people, because AI scales dysfunction instantly. Fortunately, we can learn from the past:<\/p>\n

Agent workforces and human workforces fail in the same way.<\/strong><\/p>\n

Understanding the 7 major parallels between the two will unlock the next trillion dollars of AI value creation:<\/p>\n

\n
\n
\"\"<\/picture><\/div>\n

 <\/figure>\n<\/div>\n

The tokenmaxxing hype cycle ran its full course in under a month.<\/p>\n

But the amount of tokens spent was never the real problem.<\/p>\n

People are spending so much on tokens because they don\u2019t know how to use them.<\/strong><\/p>\n

Maybe 1 in 100 employees knows how to give AI context. It\u2019s a rare breed of person that can articulate a process clearly, that has the patience to empathize with a polluted context window, or even understands what that means.<\/p>\n

Give an agent harness to the other 99 people and they will produce \u201cloops\u201d.<\/p>\n

In Claude Code\/Cowork, Copilot, Karpathy\u2019s Autoresearch, or any harness, loops are a bandaid for the fact that almost nobody can successfully prompt.<\/p>\n

Loops are a brute force attempt to compensate for human inadequacy. Agents call themselves to fix themselves only because a human never articulated the task cleanly. Brute force becomes the system\u2019s only path to progress. This all stems from human failure to successfully understand the task in the first place.<\/p>\n

You are spending tokens on spending tokens.<\/strong><\/p>\n

\n
\n
\"\"<\/picture><\/div>\n

 <\/figure>\n<\/div>\n

Most companies today are mismanaged.<\/p>\n

The vast majority of workers don\u2019t meaningfully impact the business at all. They are cogs in the machine, stamping approvals at each layer and hiring more cogs to fuel a machine that exists to exist.<\/p>\n

They\u2019re looping.<\/p>\n

Often it\u2019s more efficient to cut the loop. Elon cut 80% of X\u2019s staff and the company performed better. Private equity operating partners make a living arbitraging this simple fact.<\/p>\n

Just like 80% of employees do nothing, 80% of tokens today do nothing.<\/strong><\/p>\n

People create more people. Tokens create more tokens. Looping is the new empire building.<\/p>\n

\n
\n
\"\"<\/picture><\/div>\n

 <\/figure>\n<\/div>\n

The promise of software was that we\u2019d build it once, run it forever at low cost, and never need to supervise it. AI broke that promise. As soon as software could do anything, it couldn\u2019t do any one thing predictably.<\/p>\n

Tokens behave like a workforce, and as soon as you see tokens as employees, the promises of AI start to break down:<\/p>\n