Crypto's real momentum isn't in the charts; it's in developer activity
Opinion by: Markus Levin, co-founder of XYOThe crypto community often experiences periods of heightened anxiety. Market downturns are often triggered by counterproductive sentiment-driven events rather than by fundamental issues, creating a significant disconnect between price behavior and the actual progress being made within the industry by the companies within it. What often goes unnoticed is how much real development happens during these downturns. While market movements capture most of the attention, teams are building faster and more deliberately behind the scenes than ever. The focus shifts away from price speculation and toward real execution. Growth happens during downturns. It's a necessary phase for projects that thrive in a volatile industry. They re-focus attention on refining their technology and business, fueling the next wave of progress.As a result, there's a disconnect between online sentiment and conversations between blockchain industry leaders. For builders and project leaders, the atmosphere is of determination, not doom.Regulators are coming on boardOne of the most promising developments is the accelerating momentum of regulation policy. Many European companies are applying for MiCA licenses in preparation for regulatory updates. There's also a significant policy shift under new US leadership as the SEC retreats from several high-profile crypto enforcement actions. The disparity between sentiment and reality serves as a reminder that price is a lagging indicator. Selloffs are triggered by uncertainty around tariff announcements and background activity such as interest rates. Material, long-term statistics speak for the virtually universal optimism among industry leaders as the number of active developers has remained stable, and the number of established developers almost doubled last year. That's an incredible jump in only one year. From hype to substanceMaturation means teams thoughtfully building, governments engaging seriously with legislation, and users demanding better UX and real utility. The industry has a well-established pattern — market corrections wipe away hype and encourage focus. The last bear market gave rise to breakthroughs in DeFi, NFTs, and zero-knowledge tech. This time, it's about real-world infrastructure, regulation-ready platforms, and next-gen scalability.What emerges in these periods tends to be less visible but more durable. Teams that remain active are often those with clear models, sufficient runway, and a willingness to adapt. These are the periods when we learn whether the systems being built can handle real-world demands. One of the most promising frontiers lies at the intersection of AI and blockchain, the most ubiquitous being within Large Language Models. AI is, however, only as good as the data it's trained on.AI systems are evolving rapidly, but their foundations are skewed. They're built primarily on data scraped from the digital-first countries that predominantly lie in the northern hemisphere, which dominates global media production and internet usage. This creates a feedback loop where Western and East Asian perspectives and widely spoken languages such as English and Mandarin are not only amplified but leave little room for necessary data from smaller populations.A report from Web3 Technologies said 60% of tier-one media on the internet is English. Prominent among these media outlets is The New York Times, which has sued OpenAI based on copyright infringement. The publication alleges that their copyright-protected data was used to train OpenAI's LLM model. Recent: The future of finance is built on Bitcoin — Ethereum was just the testnetKnowing the full extent of the global imbalance in the data creating AI outputs is impossible. Allegations like this and the results delivered when using AI tools suggest the pressing need for a solution.It's even worse. When AI systems are trained on narrow, incomplete data sets, the results can exclude billions from the benefits of emerging technologies. As IBM highlights, data bias isn't just a technical issue — it's a human one with real-world consequences in healthcare, finance, agriculture, and beyond.It's become normal to use AI data every day. We receive personalized Google search results, Adobe has built AI into its industry-standard graphic and video software, and we use AI assistants like Gemini, Grok, and ChatGPT to formulate the thoughts with which we represent ourselves. All of these tools are affected by an overwhelming bias toward the center of a bell curve within their data sets, unable to access or address less common use cases.A popular example demonstrates this issue: Until recently, image generators could not create a full wine glass. No matter what prompt you provided, a wine glass full to the edge was beyond the capabilities of all known generative AI software because they had never been provided photos of wine glasses full to the brim. Their data sets had to be updated to correct this co
Opinion by: Markus Levin, co-founder of XYO
The crypto community often experiences periods of heightened anxiety. Market downturns are often triggered by counterproductive sentiment-driven events rather than by fundamental issues, creating a significant disconnect between price behavior and the actual progress being made within the industry by the companies within it. What often goes unnoticed is how much real development happens during these downturns. While market movements capture most of the attention, teams are building faster and more deliberately behind the scenes than ever. The focus shifts away from price speculation and toward real execution. Growth happens during downturns. It's a necessary phase for projects that thrive in a volatile industry. They re-focus attention on refining their technology and business, fueling the next wave of progress.
As a result, there's a disconnect between online sentiment and conversations between blockchain industry leaders. For builders and project leaders, the atmosphere is of determination, not doom.
Regulators are coming on board
One of the most promising developments is the accelerating momentum of regulation policy. Many European companies are applying for MiCA licenses in preparation for regulatory updates. There's also a significant policy shift under new US leadership as the SEC retreats from several high-profile crypto enforcement actions.
The disparity between sentiment and reality serves as a reminder that price is a lagging indicator. Selloffs are triggered by uncertainty around tariff announcements and background activity such as interest rates. Material, long-term statistics speak for the virtually universal optimism among industry leaders as the number of active developers has remained stable, and the number of established developers almost doubled last year. That's an incredible jump in only one year.
From hype to substance
Maturation means teams thoughtfully building, governments engaging seriously with legislation, and users demanding better UX and real utility. The industry has a well-established pattern — market corrections wipe away hype and encourage focus. The last bear market gave rise to breakthroughs in DeFi, NFTs, and zero-knowledge tech. This time, it's about real-world infrastructure, regulation-ready platforms, and next-gen scalability.
What emerges in these periods tends to be less visible but more durable. Teams that remain active are often those with clear models, sufficient runway, and a willingness to adapt. These are the periods when we learn whether the systems being built can handle real-world demands. One of the most promising frontiers lies at the intersection of AI and blockchain, the most ubiquitous being within Large Language Models. AI is, however, only as good as the data it's trained on.
AI systems are evolving rapidly, but their foundations are skewed. They're built primarily on data scraped from the digital-first countries that predominantly lie in the northern hemisphere, which dominates global media production and internet usage. This creates a feedback loop where Western and East Asian perspectives and widely spoken languages such as English and Mandarin are not only amplified but leave little room for necessary data from smaller populations.
A report from Web3 Technologies said 60% of tier-one media on the internet is English. Prominent among these media outlets is The New York Times, which has sued OpenAI based on copyright infringement. The publication alleges that their copyright-protected data was used to train OpenAI's LLM model.
Recent: The future of finance is built on Bitcoin — Ethereum was just the testnet
Knowing the full extent of the global imbalance in the data creating AI outputs is impossible. Allegations like this and the results delivered when using AI tools suggest the pressing need for a solution.
It's even worse. When AI systems are trained on narrow, incomplete data sets, the results can exclude billions from the benefits of emerging technologies. As IBM highlights, data bias isn't just a technical issue — it's a human one with real-world consequences in healthcare, finance, agriculture, and beyond.
It's become normal to use AI data every day. We receive personalized Google search results, Adobe has built AI into its industry-standard graphic and video software, and we use AI assistants like Gemini, Grok, and ChatGPT to formulate the thoughts with which we represent ourselves. All of these tools are affected by an overwhelming bias toward the center of a bell curve within their data sets, unable to access or address less common use cases.
A popular example demonstrates this issue: Until recently, image generators could not create a full wine glass. No matter what prompt you provided, a wine glass full to the edge was beyond the capabilities of all known generative AI software because they had never been provided photos of wine glasses full to the brim. Their data sets had to be updated to correct this comical problem, which revealed a much more serious one.
Decentralized data offers a solution. Globally incentivized systems like DePINs enable the participation of populations that would otherwise remain underserved, allowing the valuable data they provide to come online. This improves the service for everyone, making smaller global communities more accessible to commerce and enabling them easier access to the rest of the world. It also empowers smaller data creators to monetize their data rather than relinquishing it to tech giants.
Where do we go from here?
The crypto industry is entering a new phase. A phase that's more productive and sustainable. Expect to see rapid growth in working infrastructure, platforms and applications that welcome knowledgeable, consumer-friendly regulations and projects that respect the time and money of their users.
Opportunities within the crypto space are changing but not shrinking. Our opportunities grow as we learn from what has not worked in the last few years. They will take time to develop, but successful builders will focus on long-term, incremental change and sound business practices rather than chasing fads and short-term profits.
The momentum of real progress has never been stronger, and it is precisely during times like these, when it feels like no one's watching, that the foundations of the future are laid.
Opinion by: Markus Levin, co-founder of XYO.
This article is for general information purposes and is not intended to be and should not be taken as legal or investment advice. The views, thoughts, and opinions expressed here are the author’s alone and do not necessarily reflect or represent the views and opinions of Cointelegraph.
