It’s 2pm on a Tuesday. You were supposed to be working on your important project. Instead, you’ve been testing the latest AI tool you discovered on LinkedIn for the past three hours. “Just trying it out,” you thought. Now you’ve spent $50 on credits, lost the afternoon and the project is still on hold.

This scene is repeated every day in countless offices and home offices. AI tools promise to make us more productive. But for many people, they become the biggest obstacle to productivity. What starts as harmless experimentation quickly turns into a time and money pit that costs both individuals and companies millions.

As the author of this article, I must admit that I, too, have fallen into this trap. This month, I was shocked to discover that I had invested a large four-digit sum in AI tools. The many small investments — 20 euros for credits here, 50 euros for a premium subscription there — added up to an eye-opening total at the end of the month.

Parallels with gambling

This experience painfully reminded me of my time as a World of Warcraft player. Back then, I spent all my time in a parallel world of farming, raids and endless quests. The monthly fees kept me hooked on the game. To make sure I was always well-equipped, I even paid so-called “China farmer bots” to supply my character with virtual gold or level up my characters while I was working or sleeping. I completely lost touch with money and time – luckily, I realized this in time.

The parallels between then and now are frightening. Both times, it was small, seemingly harmless expenses. Both times, I lost track of time and had the illusion that I was doing something productive when I was actually just wasting time and money. The only difference is that today, the addiction is disguised as “professional development” and “preparation for the future.”

The problem is complex. On the one hand, we want to “stay on the ball” in the face of rapid AI development. On the other hand, constantly trying out new tools leads to a modern form of procrastination that hides behind the guise of “further education.” One user describes the AI image generator Midjourney on Reddit: “.”

The parallels to gambling are no coincidence. AI companies use the same psychological tricks as casinos:

• Low entry costs
• Unpredictable rewards
• The promise of a big breakthrough.

Initial studies confirm the consequences. For example, a found that ChatGPT power users develop withdrawal symptoms, think about AI constantly and neglect real work and relationships. What is sold as a productivity boost often turns out to be the opposite — a digital distraction that eats up valuable time and a lot of money.

The procrastination paradox

The modern working world presents us with a paradox: we are so preoccupied with productivity tools that we are no longer productive.

This phenomenon even has a name: “tool hopping” or “shiny object syndrome.” Every day, new AI applications appear on the market. Product Hunt, LinkedIn and tech blogs bombard us with promises of revolutionary breakthroughs. “This AI will transform your business,” “The tool that changes everything,” “Finally: AI that really works” — headlines like these trigger our desire for innovation and, at the same time, our fear of missing out.

The problem is exacerbated by the low barriers to entry. Most AI providers offer free trials or inexpensive starter packages for their tools. “Just take a quick look” doesn’t cost anything. But this apparent harmlessness is deceptive. Time is money, and the hours we spend testing new tools are hours we could be spending on our actual work.

A typical example: A marketing manager discovers a new AI tool for social media posts. He invests two hours to understand it, another three hours to set it up and another four for initial testing. At the end of the day, they have invested nine hours in a tool that they may never use again.

What’s particularly insidious is that many people feel productive even though they are not. Psychologists call this “pseudo-productivity” — the illusion of doing something important when you are actually just wasting time.

Credits as a cost trap

The financing of AI tools follows a system that is no less perfidious, concealing the true costs. Similar to how casinos work with chips, AI providers use credits, tokens or points. This abstraction from real money lowers the psychological barrier and leads users to spend more than they originally planned.

A typical scenario: You buy 1,000 credits for $20. That sounds like a fair deal. But you quickly discover that more complex requests consume more credits than expected. A high-quality image costs 50 credits, a longer text 100. After a few hours, the credits are used up, but you haven’t achieved the desired result. So you buy more — another $20, then another $20.

This microtransaction strategy comes from the gaming industry and is psychologically highly effective. Small amounts feel harmless, but they quickly add up to significant sums. Many users spend more on AI tools each month than on their streaming subscriptions — without even realizing it.

What’s particularly problematic is that pricing is often opaque. How much does image generation really cost? How many tokens does a complex text query use? Most providers deliberately make this information difficult to access. Users buy credits “on spec” and hope that they will be enough.

The cost trap is exacerbated by the variable quality of the results. Not all AI output is usable. Often, you have to make several attempts before you get the right result. But each attempt costs credits. A single usable image can quickly cost 200 to 300 credits, even though the provider advertises “only 10 credits per image.”

A freelance designer reports: “I just wanted to try Midjourney. At the end of the month, I had spent $180 on images that I couldn’t use for the most part. My Adobe Creative Suite subscription is cheaper.”

Companies are particularly affected by this: if several employees test different AI tools at the same time, the costs can quickly run into four figures.

The dopamine cycle

Our brains are not designed for the digital world. The mechanisms that helped us survive are now being used against us.

The key lies in dopamine: contrary to popular belief, this hormone does not make us happy — it makes us hungry for more. That is why dopamine is not released when we receive a reward, but when we expect it. The moment before clicking “Generate” is more exciting than the result itself.

Scientists have found in a study that . If you never know when you’re going to win, the excitement stays high. AI tools use exactly this principle. You never know whether the next image generated will be brilliant or just mediocre. This uncertainty keeps you hooked.

The problem with this is that most of the results are mediocre, like the small wins on a slot machine. But now and then, you get a “jackpot” experience. A perfect image. A text that fits exactly. These rare hits motivate us to keep going.

The humanization of AI exacerbates the situation. Chatbots such as ChatGPT are programmed to appear as human as possible. They use emoticons, ask questions and seem to show understanding. This makes them even more appealing. People build relationships with the systems as if they were real people.

A study by OpenAI and MIT confirms: People who use ChatGPT intensively develop an emotional attachment to AI. They think about the technology constantly, feel uncomfortable when they can’t use it and neglect real relationships.

People who are lonely or dissatisfied with their work are particularly at risk. They seek refuge from reality in AI tools. AI does not judge, is always available and always provides feedback. This can lead to a parasocial relationship — a one-sided, emotional attachment to an artificial entity.

Staying productive in the AI era

The solution is not to avoid AI tools altogether. The technology offers enormous potential for productivity and creativity. The problem lies in how we use it. With the right strategy, you can enjoy the benefits without falling into the time and cost trap. Here are a few tips on how to do this:

• Set clear boundaries. Decide in advance how much time and money you want to spend on AI experiments. Treat it like a hobby budget — with fixed limits that you don’t exceed.
• Distinguish between productive use and procrastination. Before each AI session, ask yourself: “Am I solving a specific problem, or am I just experimenting out of curiosity?” If you don’t have a clear goal, leave it alone.
• Schedule fixed times for AI experiments. For example, one hour on Friday afternoons. Outside these times, AI tools should be off limits. Use timers and apps that limit your screen time to help you stay focused. Your brain needs time to process the results.
• Focus on a few proven tools instead of constantly trying out new ones. Become an expert in one tool before testing the next. Quality over quantity is the key.
• Measure the actual benefits. Keep a log of the time and money spent on AI tools. Document the specific results you have achieved with them. You will be surprised how often the cost-benefit ratio is negative.
• For companies, develop a central AI strategy. Instead of letting every employee experiment wildly, a few experts should evaluate new tools and provide proven solutions for the entire team. This saves time, money and prevents chaos.
• Learn to be satisfied with mediocre results. Not every AI result has to be perfect. Often, “good enough” is good enough. Perfectionism is a driver of excessive use. Accept that AI tools are tools, not miracle machines.
• Diversify your skills. Don’t rely solely on AI for creativity or problem-solving. Use traditional methods as well. Draw with pen and paper. Write without AI assistance. Solve problems by talking to real people. This will sharpen your skills and reduce your dependence.
• Be especially careful with chatbots. Treat them as tools, not “friends.” Don’t tell bots intimate details. Seek help from real people — friends, family or therapists — for emotional problems. AI can inform and inspire, but it cannot replace real relationships.

AI tools are here to stay. The future belongs to those who can use AI intelligently. That doesn’t mean spending as much time as possible with the tools, but using them at the right time for the right purposes. Quality over quantity, strategy over chaos, productivity over procrastination – these are the keys to successful AI use.

This article is published as part of the Foundry Expert Contributor Network.
Want to join?