Amazon Web Services (AWS) and Microsoft are pulling back from AI data center investments, suggesting the issue of centralized models. Analysts have made this latest development and have reiterated why decentralized blockchain-based infrastructure is the solution.
Kai Wawrzinek, co-founder of Impossible Cloud Network, discussed these looming questions in an exclusive interview with Beincrypto.
AI data center crashed into a wall
A few months ago, AI seemed like one of the most promising areas in the global high-tech industry. However, this photo looks very different as companies like AWS and Microsoft have announced a pause in building AI data centers. what happened? What does the future of AI look like? Kai Wawrzinek explained the situation today:
“The news that AWS is joining Microsoft to withdraw from new data centers when AI is growing exponentially is evidence of the enormous inefficiency this model presents to expand the global internet. Microsoft and AWS may recognize that the intensive infrastructure model cannot adapt quickly enough,” Wawrzinek argued.
AWS and Microsoft are not the only companies facing these issues. Meta publicly claimed that it would spend hundreds of billions on AI infrastructure and data centers, but in less than three months it called for it to fund its competitors.
Openai was also shaken by the enormous cost of operating ChatGpt. Sam Altman tacitly admitted that the study might never be useful.
Wawrzinek is looking for a clear solution. We abandon the intensive model completely and focus on Defai. These industry leaders have accumulated billions of dollars in CAPEX and pioneered LLM development, but the overall strategy could be self-destructive.
For example, the construction of American AI data centers has moved electrical engineers to an unprecedented degree. So many experts focus on the centre itself, creating bottlenecks for skilled labor.
This harms renewable energy projects and electrical grids, and ironically harms the functioning of data centers.
“In the AI age, infrastructure is needed to suit its speed and scale, and distributed systems are the only model built for its future. In contrast, a decentralized, market-driven approach can solve this problem.
Can defai handle the issues?
Compared to a centralized data center model, defai has improved AI computational accessibility. Blockchain-enabled economic incentives can accelerate deployment speed, improve scalability, and optimize resource allocation without large-scale leading capital.
In short, these distributed systems are more agile than their competitors.
Blockchain-based AI companies have been able to leverage important computing power without a centralized data center. For example, Depin Firm Aethir has made great strides in the GPU-AS-A-Service model.
Other companies, like 0G Labs, have proven that decentralized development of AI development is theoretically not viable. It is beneficial and necessary for the ecosystem.
If this all seems exaggerated or utopian, it’s important to remember Deepseek, the AI’s “black swan” event.
The genai model, which puts the Chinese market, has proven worldwide that AI companies can create cutting-edge LLMs at a fraction of their hardware costs. Therefore, the AI industry may need to completely rethink its data center model if this one developer proves to be very successful.
Skeptics wonder whether decentralized AI can compete with data centers, but the reality is that centralization has its own inefficiencies.
“The future of AI infrastructure lies in open, unauthorized networks where supply meets demand dynamically and globally.
So far, centralized AI companies have accumulated billions of venture capital investments, but their ability to innovate has hit a brick wall. You may need a better model to produce the best possible results.
