Jake Brukhman argues that AI development is becoming increasingly centralized, making advanced models more vulnerable to control and regulation. Distributed AI projects are exploring ways to
There has been continued scrutiny over who controls advanced AI models intensified after Anthropic complied with U.S. export restrictions, prompting renewed discussion around decentralized AI networks. CoinFund founder Jake Brukhman argued that the incident highlights how AI development is becoming increasingly centralized. He said distributed AI networks could offer an alternative by using global GPU resources to train frontier models outside concentrated infrastructure.
According to Brukhman, AI models have long represented a centralizing force and a natural target for government oversight. He pointed to Anthropic’s recent compliance with U.S. export controls as evidence of that trend.
Anthropic suspended access to Fable 5 and Mythos 5 after receiving a directive affecting foreign nationals. The company said the restrictions applied regardless of location, leading it to disable both models.
Against that backdrop, Brukhman argued that decentralized networks can provide a counterweight. He said the industry’s biggest challenge is not demand for AI, but access to large-scale computing power.
Moving beyond the policy debate, Brukhman said enough commodity GPU capacity exists worldwide to compete with centralized AI systems. However, he noted that distributed training requires new algorithms capable of coordinating those resources efficiently.
According to Brukhman, several teams are actively researching that problem. He named Gensyn, Prime Intellect, Bagel, Pluralis Research, Nous Research, Macrocosmos, and Covenant among projects pursuing distributed AI training.
He said their work demonstrates that decentralized training can be both feasible and cost-efficient. CoinFund has also backed several companies in the sector. Prime Intellect secured seed funding co-led by CoinFund in 2024 before raising additional capital later.
Beyond computing infrastructure, Brukhman highlighted another challenge facing open-source AI projects: sustainable economics.
He said open models often lack a durable business model despite broad accessibility. According to Brukhman, Pluralis Research is exploring a different approach by dividing model weights among network participants.
He argued that this structure could support tokenized AI models while preventing any single participant from controlling the entire system.
Meanwhile, Brukhman said the broader discussion now centers on whether AI development remains concentrated among a few organizations or expands through open decentralized networks.
The post CoinFund Founder Jake Brukhman Warns of Growing AI Control Risks appears on Crypto Front News. Visit our website to read more interesting articles about cryptocurrency, blockchain technology, and digital assets.
