A sharp fall in the cost of artificial intelligence has reignited debate about the future of work, power and governance, after a social media post drew attention to what one user described as a “civilization-scale” shift in how intelligence is produced and deployed.

The discussion was sparked by Aakash Gupta, a social media user, who posted on X (formerly Twitter) about the dramatic collapse in AI prices over the past two years and its implications for the global economy and labor markets.

“GPT-4 launched at $60 per million output tokens. Today, equivalent capacity costs less than $1. That’s a 98% price collapse in two years. Demand didn’t fall. It exploded,” Gupta wrote. He noted that OpenAI’s annual recurring revenue (ARR) rose from $1 billion to more than $12 billion, even as prices fell quarter after quarter.

Gupta framed the change through the lens of the Jevons paradox, an economic principle that holds that efficiency gains often lead to higher overall consumption than conservation. Drawing a historical parallel, he wrote: “When coal became cheap in the 1800s, England didn’t use less coal. They burned 10 times more. Intelligence follows the same curve, except that the rate of adoption is compressing a century of energy economy into 36 months.”

According to Gupta, the consequences extend far beyond automation. Citing Stanford research, he highlighted a 280-fold reduction in AI computing costs between 2022 and 2024, noting that tasks that once cost $1,000 can now be completed for less than $4.

“At this price, companies aren’t just automating what humans used to do. They’re starting to do things that were never economically viable at human labor prices,” he said, adding that analytical work that once required a highly paid analyst for a year can now be completed “for $50 in an afternoon.”

As intelligence becomes abundant and cheap, Gupta argued, it ceases to be the scarce input. “Taste, judgment and the ability to ask the right question become the bottleneck,” he wrote. “Returns flow to people who can direct the intelligence, not to people who provide it.”

Gupta’s post was shared alongside another user’s claim that “There is an unlimited demand for intelligence,” a claim that drew agreement and concern across the platform.

One response warned that the central risk lies not in abundance, but in uncontrolled scale. “‘Unlimited demand for intelligence’ is true. The dangerous part is what people conclude from it,” the user wrote. “When intelligence becomes cheap, the scarce resource is not ‘taste’. It’s stability.”

The commenter warned that cheap AI not only replaces labor, but scales decisions, often faster than institutions can regulate them. “When coal got cheap, we didn’t just burn more coal. We built machines that could burn it faster than we could regulate the consequences. AI is that, but for cognition,” the post said.

The real bottleneck, the user argued, is the absence of safeguards, citing the need for reliability, auditability, application of restrictions, provenance, contradictory evidence, rollback mechanisms, human readability and institutional accountability.

“Otherwise, you’ll get a world where every company manages 10,000 autonomous analysts, sending decisions to production at machine speed, with no consistent oversight,” the user added. “Cheap intelligence is not the end. Regulated intelligence is.”

A third voice questioned whether demand for intelligence would keep pace with its rapidly expanding supply, suggesting the trend could deepen inequality.

“I think the current trajectory is to reach the point of an almost infinite supply of intelligence,” the user wrote. “Demand will not grow at the same rate as supply and will likely redefine the labor market.”

The commentator predicted a more polarized society, where individuals and companies able to effectively harness AI reap disproportionate gains, while others are left behind. Despite the dramatic reductions in unit costs, they argued that “the average user of the AI ​​platform is a long way from being able to increase their productivity/output in a meaningful way.”

What is Jevons’ paradox?

Jevons’ paradox holds that when a resource becomes cheaper or more efficient to use, total consumption often rises rather than falls. First noted in the 19th century by British economist William Stanley Jevons, the phenomenon arose when more efficient steam engines led Britain to burn much more coal, not less. Cheaper energy made new applications viable, expanding demand in industry, transportation and manufacturing.

The same logic is now being applied beyond energy, especially to AI and computing. As AI becomes cheaper, faster and more accessible, it not only replaces existing tasks, but enables entirely new uses that were previously uneconomical. That, advocates argue, is why falling AI costs may fuel an explosion in demand for intelligence rather than curb it.