It's like having your cake and eating it too.

NEAR, a blockchain network built by one of the original architects of modern AI, said Wednesday that people can now pay for artificial-intelligence usage by staking the project's native token, an alternative to more common methods of covering the costs like getting billed per request or signing up for a subscription.

Staking is where blockchain users deposit tokens into a protocol, and get a yield as the reward – a way of securing transactions on the network.

Under NEAR's new setup, the staking yield is converted into monthly credits that will go toward covering the cost of AI computation, as laid out in a blog post. If users decide to discontinue, they can simply unstake the tokens.

"You never lose your principal, and you don't need to pay out of pocket," NEAR founder Illia Polosukhin told DeAI News Thursday in an exclusive interview at the Blockchain Application Stanford Summit (BASS) in Palo Alto, California. "It's fully non-custodial, fully decentralized."

The new payment method comes as more users and companies wrangle with the fast-rising costs of artificial intelligence, and it provides fresh utility for cryptocurrencies beyond price speculation or yield harvesting.

NEAR's offering resembles a similar mechanism previously rolled out by Morpheus, another decentralized AI project, whose "marketplace API" allows users to redirect staking yield to pay for AI computation. (In May, Distro Media, publisher of DeAI News, built a prototype of an editorial agent called JournoBot that funds its own inference with roughly $0.15 a day in rewards from 50 MOR staked on the Morpheus decentralized-compute network, as previously reported.)

Such staking mechanisms could also dovetail with a growing industry preference for open-source AI software models that are typically free to download and inspect. Users still must cover the costs of running their AI jobs on cloud computing, often via gateways like OpenRouter, which require an account.

NEAR said in the blog post that the new service, called Staking for NEAR AI, opens access to all 43 models on its platform, including ones from Anthropic, OpenAI and Google.

The 'T' in GPT

Polosukhin is seen as a prominent leader in decentralized AI because he worked on artificial-intelligence research before he turned his efforts to blockchain. He co-authored the 2017 Google paper "Attention Is All You Need" that introduced the Transformer architecture underpinning today's large language models.

He has warned that the current trajectory of AI could hand a handful of corporations the power of "mass manipulation" over how billions of people perceive reality, as reported by TheStreet in February.

On July 14, NEAR made waves when it joined the x402 Foundation, a Coinbase-hatched, now Linux Foundation-stewarded standard for agent payments whose backers include Visa, Mastercard and Stripe, as DeAI News reported.

The move was seen as a sign of NEAR's endorsement of that payment technology, which allows AI agents to easily route payments over blockchains for data, content or even the costs of AI processing.

Impact on NEAR supply

NEAR noted that any extra demand for its tokens from AI staking could have the effect of tightening the supply – typically seen as a source of upward price pressure.

NEAR's token price has slumped 33% this year, though it still has a market valuation of $2.2 billion, according to CoinGecko.

"As more of the network's AI is paid for this way, more of the supply is committed to work being done, and the value that activity creates routes back toward the people who hold and secure the network instead of pooling at a platform above them," the blog post reads.

(HOW AI WAS USED IN THE PRODUCTION OF THIS PIECE: The first draft of this story was written on Claude Desktop using a custom blog-rewrite skill, supplemented with background research on NEAR and its co-founder and with reference to prior DeAI News coverage. I then used our Distro Publisher MCP connector to file the draft directly to DistroVerse, our web app. I used the DistroVerse native editing interface to edit the story and add the exclusive quotes from the Polosukhin interview. I checked the facts against the original blog post and the cited sources before publication.)