Tech startups have a new dream: companies worth billions, but needing hardly any people. Behind this is the trend of **Tokenmaxxing**, a strategy involving massive investment in AI tokens. The goal is to automate business processes to such an extent that human labor becomes almost superfluous.
This hype promises incredible efficiency, but also carries **enormous risks**. It not only shifts power dynamics in the economy but could also create a new kind of cost trap. If companies blindly invest in expensive AI tokens without checking their true value, disappointment and high expenses that will never be recouped are likely. This is not guaranteed success, but a **high-risk experiment**.
A recent report from Golem.de highlights how tech startups are pouring ever-larger sums into **AI tokens**. These tokens are the 'currency' with which AI language models like ChatGPT or Claude process their computing power and tasks. The more tokens, the more 'work' the AI can do. The vision: companies that achieve billions in revenue with minimal staff – or none at all.
For you as an employee or freelancer, this could mean that routine tasks are increasingly taken over by AI. Repetitive jobs are at stake. At the same time, new professions are emerging: as an **AI architect** or **prompt engineer**, who controls the automation. Your ability to work with and understand AI will be crucial for your future.
Companies, especially small and medium-sized ones, face a tough challenge. Those who don't keep up risk falling behind. But those who blindly invest in 'Tokenmaxxing' could burn a lot of money. It's about carefully examining which processes are truly worth automating and where the **human-in-the-loop** – meaning human oversight – remains indispensable. Otherwise, high licensing costs without real added value and a **lock-in** with expensive AI providers are a threat.
The biggest opportunity lies in **radical efficiency gains**. Processes that currently take hours or days could run in minutes. Companies could react faster to market changes and develop entirely new business models. This means a potential boost for innovation and the ability to offer products and services at prices previously unthinkable.
The biggest risk is the **exploding costs for AI tokens**. If a startup builds its business model on this expensive resource, any price change by major AI providers can upset the entire calculation. Furthermore, the complexity of automation is enormous. AI systems can hallucinate – meaning they give **invented answers** – or make mistakes that become costly without human control. Dependence on a few large AI providers also creates a dangerous **vendor lock-in**.
