I recently read a paper titled 'Artificial Intelligence Layoff Trap' ， Trembling with fear, this paper, jointly published by two researchers from the University of Pennsylvania and Boston University, should be able to keep CEOs who are crazily implementing AI layoffs awake. The core logic is actually very simple. Big companies replace manual labor with AI, which seems to save costs, but at the same time, they also eliminate their future customers. The laid-off employees no longer consume, and the more people are laid off, the weaker the overall market consumption ability. In the end, the enterprise will find that the product has been made, but no one can afford it anymore. This principle is actually understood by every corporate executive, and the economic logic is not complicated. But the question arises, why hasn't anyone stopped? The paper describes the prisoner's dilemma: If you don't do automation, your competitors will They reduce costs, lower prices, and seize market share If you don't take action, you'll just wait to be killed So even if every company knows about the collective destruction, they can only grit their teeth on it Researchers have demonstrated that this is a real-time prisoner's dilemma game. Let's take a look at the actual data listed ， Just know how crazy it is: Block has laid off nearly half of its employees this year (nearly 5000 people), and Jack Dorsey directly stated that AI has made these positions unnecessary. It is expected that most companies will come to the same conclusion next year. Salesforce replaced 4000 customer service representatives with AI. Goldman Sachs has deployed programming tools, allowing one engineer to handle the work of five people. By 2025, over 100000 technology professionals will be laid off, with more than half of them clearly attributed to AI. 80% of jobs and tasks in the United States can be automated by AI. The research team tested all common solutions, but ultimately all went bankrupt! Universal Basic Income (UBI)? Do not change the incentives for automation in any company Capital gains tax? Adjusting the profit level cannot change the decision to replace personnel at the individual task level Collective bargaining? Cannot be sustained, because automation is always the dominant strategy What's even more terrifying is that they also discovered a 'Red Queen Effect'. Better AI will not solve problems, it will only accelerate them. Every company pursues faster automation to grab market share, but in the end, everyone becomes automated and their advantages cancel each other out, leaving only destroyed demand. The only possible effective solution proposed by the study is to levy a Pigouvian tax. Automated tax charged by task. Force the company to pay for the demand destroyed by each replacement of workers. From an economic perspective, this will trigger a more deadly chain reaction, a deflationary spiral. Let's take a look at this logical chain: Step 1: Decrease in salary income AI replaces workers, leading to an increase in unemployment rate On the job employees are worried about being laid off and dare not ask for a raise The total wage income of the whole society continues to decline Step 2: Collapse of Effective Demand Keynes had long said that the core of economic growth is effective demand • Decreased wages=Decreased purchasing power=Enterprises unable to sell and ship • Decreased corporate revenue, further layoffs to reduce costs Step 3: Price wage deflationary spiral Enterprises are forced to lower prices in order to reduce inventory Price reduction leads to shrinking profits and continued layoffs Wages further decline and consumption becomes more sluggish • Form a self reinforcing death spiral Step 4: Overcapacity and Liquidity Trap Production efficiency has greatly improved due to AI, but no one is buying it Enterprises are unwilling to expand production even if they borrow money at zero interest rates The central bank's interest rate cuts have failed, and the transmission mechanism of monetary policy has broken The economy has been in a Japanese style slump for 30 years The terrifying aspect of this logical chain is that traditional economic tools are almost completely ineffective: • Interest rate cuts? Enterprises do not lack money, what they lack is demand Fiscal stimulus? The government sends money, but companies continue to lay off employees, and the money cannot flow to the entity • Printing money with water? Money has all entered the asset market, and ordinary people cannot afford to buy things Trade war? Deflation is happening globally, and protectionism will only accelerate the decline The above situation is very similar to the current Chinese economy. We can observe the current domestic economy to understand. There is severe overcapacity in the manufacturing industry, with continuous negative growth in PPI and CPI approaching zero. It's not that they don't want to stimulate consumption, it's that ordinary people don't have money in their pockets and companies are still reducing costs and laying off employees. This is a typical deflationary situation. And AI is accelerating the global journey towards this process: Unlike previous technological revolutions that created new jobs, AI directly replaces mental labor The replaced white-collar workers are the main force of consumption, and the collapse of their consumption ability has a huge impact The speed at which AI improves efficiency far exceeds the speed at which new demands are created • Final result: overproduction+insufficient demand=long-term deflation The only similar scene in history was the Great Depression of 1929. At that time, there was also an increase in production efficiency due to the electrification revolution, but the wage growth of workers could not keep up, and ultimately demand collapsed. It took 10 years for Roosevelt's New Deal to recover, and in the end, it was only through the demands of World War II that it truly emerged. This AI revolution, reminded in the paper ⚠️， Perhaps more dangerous than the Great Depression. Because: • Faster replacement speed and wider range Globalization makes it impossible for countries to be self-sufficient The central bank's toolbox has been emptied by 2008 and 2020 The debt level is much higher than in 1929, and fiscal space is limited The final conclusion is sobering: this AI revolution is not simply about transferring wealth from workers to bosses. Both sides lose. Workers lose their income, and companies lose their customers. This is a senseless loss that market mechanisms themselves cannot prevent. What's even more alarming is that this will trigger systemic economic deflation, and once deflation forms expectations, it's like a black hole that's difficult to reverse. The lesson of Japan is right in front of us. I think this logic is also applicable in the encryption industry. When AI begins to replace developers, operators, and community management on a large scale, Web3 project developers may also need to consider whether their token economy model can withstand this wave of demand collapse? Users are out of money, who will take over the tokens from the market?