A few months ago, the crypto market experienced quite interesting dynamics from various angles — ranging from serious debates about AI safety limits to funding bubbles that sparked speculation. If you pay attention, all of it stems from three major stories currently developing.

First, there is the Anthropic drama refusing Pentagon requests to lift safety restrictions on their models. The company maintains its principles and refuses to develop technology for autonomous weapons or large-scale surveillance, even if it means losing around $200 million in government contracts. This decision triggered a chain reaction in the tech industry — even OpenAI’s CEO openly supported Anthropic’s safety stance. But the debate quickly split between those who see this as an important ethical line versus those worried about national security. One commentator even said, if America doesn’t develop this technology, other countries will.

Second is the spectacular funding news for OpenAI — $110 billion. NVIDIA invested about $30 billion, and Amazon might reach $50 billion. But here’s the issue: OpenAI’s revenue was estimated at around $13 billion last year, while cumulative losses over the next few years could exceed $115 billion. A Wall Street veteran with 45 years of experience wrote that this is the first time he’s seen the three smartest investors together put out $110 billion for a loss-making company. The community’s debate is divided: some believe this is necessary infrastructure investment for long-term AGI, others see it as a new tech bubble forming.

Third, Block announced a 40% cut with engineering team reductions reaching 70%. Jack Dorsey explained that code output per engineer increased by 40% since September of the previous year, mainly due to AI tools. This quickly sparked discussions about AI’s impact on technical jobs — whether this is proof that AI is truly transforming the job structure, or just a correction from the overhiring during the pandemic era.

On the ecosystem side, there are some interesting developments. Vitalik recently provided a specific schedule for Ethereum’s roadmap, which is rare — this is the first meeting minute where he gave a clear timeline like this. By 2026, ZK-EVM clients will start participating in network verification, and by 2027, the participation proportion will be gradually increased. The Ethereum community appears more optimistic with this signal, though there are technical concerns about concentration risks if too dependent on ZK-EVM clients.

Meanwhile, DeFi lending protocols like Morpho show much better performance compared to similar protocols — only down 39% from the peak with a 155% increase over the year. Some researchers link this to a simpler governance structure without conflicts between labs, DAOs, and core teams.

On Solana, there has been significant progress with US-licensed banks now supporting direct storage and withdrawal of Solana assets. About 13.7 million bank users can now hold SOL directly without needing to go through crypto exchanges. This is seen as a sign of deep integration between traditional finance systems and blockchain infrastructure.

The Base ecosystem is also becoming an important testing ground for integrating AI Agents and crypto applications. Some trading Agent experiments show transaction volumes reaching $4.5 million in the first hour, while other applications enable AI Agents to place bets or open positions directly within group discussions.

Overall, this period reflects a major shift: the AI industry is entering a “technology race driven by capital,” while crypto continues to deepen integration with traditional financial systems and AI technology. The big question remains whether this momentum is sustainable or if we are witnessing the formation of a larger new bubble.