A reality check for India's AI crypto rally

A surge of interest in artificial intelligence has spilled over into India’s cryptocurrency markets, where tokens tied to AI-driven blockchain projects have emerged as investor favourites.

Over the past year, the global market capitalisation of AI-related tokens has ballooned from about $2.7 billion to nearly $30 billion, underscoring how quickly investors have embraced the idea of combining two of the decade’s most-hyped technologies.

For Indian traders, these tokens offer both novelty and a hedge against the speculative frenzy of meme coins. “AI crypto” is being marketed as a way to diversify portfolios while maintaining exposure to the artificial intelligence boom that has gripped global markets.

Leading Indian exchanges such as CoinDCX, CoinSwitch, Mudrex and Giottus report that AI-linked tokens now account for a significant share of trading volumes. Among the most actively traded are Fetch.ai (FET), Bittensor (TAO), Internet Computer (ICP), Render, and NEAR.

The appeal of these assets lies in their underlying promise: decentralised AI networks that allow users to train, deploy, and monetise machine learning models collaboratively. Advocates argue that such systems offer a transparent and secure alternative to the tightly controlled AI ecosystems of global tech giants such as OpenAI and Google.

Yet, beneath the optimism, analysts warn that many of these projects remain at an early stage, with uncertain revenue models and limited practical adoption. For now, the rally appears driven more by potential rather than proof.

AI tokens trending in India

Sumit Gupta, Co-founder, CoinDCX said that in just the past one year AI token market capitalization has surged from $2.7 billion to nearly $30 billion - reflecting strong investor confidence in AI-blockchain integration.

“With AI projected to grow into a multi-trillion-dollar industry, its intersection with crypto presents promising opportunities. Leading projects like NEAR Protocol, ICP, TAO and Render are demonstrating the practical applications of AI tokens in enhancing computational power, automation, and decentralized AI services,” Gupta told Moneycontrol.

Among the country’s top two crypto exchanges, CoinDCX has seen trading volumes grow for TAO, NEAR, ICP, FET, GRT (The Graph) and INJ (Injective), while for CoinSwitch, the best performing AI tokens are AI16z (not related to a16z VC firm), Virtual, Goat, PHA (PHALA) and FET.

These five AI tokens collectively hold around 51 percent of the trading volume amongst all the AI coins for CoinSwitch. The crypto exchange has seen AI token trading volumes grow 2.5X compared to previous year.

Not just individuals, AI crypto tokens have started attracting institutional investments globally too. For instance, US-headquartered digital currency asset management firm Grayscale’s recent move to increase its holdings in TAO from 3% to 27% of its crypto-focused portfolio.

For Mudrex, the AI tokens currently make up about 15-20 percent of its overall crypto trading volumes, dominated by NEAR, Internet Computer (ICP), Artificial Superintelligence (FET), Bittensor (TAO), and Render (RNDR).

While it’s still a significant share, the exchange reported that AI crypto token trading volumes in 2025 have largely remained flat or at same levels as the hype around the AI-based blockchains have started to cool down, with investors seeking real-world use cases.

Between 2023 and 2024, trading volumes for AI tokens had jumped 1800 percent on Mudrex, driven by the heightened activity in the AI sector.

“Over the past couple of years, AI tokens have become very popular among new investors…We are seeing a trend where investors are focusing on real utility tokens rather than just tokens that use the name AI for hype. With a fast-growing base of Web3 and AI developers, the country is well-positioned to build and adopt meaningful AI-powered blockchain solutions,” said Edul Patel, CEO, Mudrex.

AI boom driving the shift

Unlike meme coins that lack fundamentals, AI tokens are backed by real-world use cases, integrating AI capabilities such as automation, predictive analytics, and fraud detection within blockchain ecosystems. They bring in the ability to autonomously execute transactions and optimize processes making them compelling use cases in areas like decentralized finance (DeFi).

“The AI and blockchain synergy has the potential to enhance efficiency, security, and decision-making in financial applications, making these tokens more than just a passing trend,” said Gupta.

CoinSwitch believes that these AI tokens are gaining steam through popularity, largely fueled by active communities. Some of the top ones are focussed on building decentralized VC DAOs (Venture Capital Decentralized Autonomous Organization), others on developing AI agents, while a few others are also operating as community-driven meme coins.

“The crypto market has always evolved through innovation cycles, first with NFTs, then Layer 2 solutions, and now AI powered projects. While still an emerging trend, the intersection of AI and crypto is gaining steady attention,” said Balaji Srihari, VP, CoinSwitch.

Further, with the soaring valuations of AI-focused companies like NVIDIA and OpenAI in the past year, crypto investors are seeing AI tokens as a bet on the AI.

The blockchain projects behind the hype

Nearly all the leading Indian exchanges have named NEAR, FET, TAO and ICP as their best performing AI crypto tokens and the reason for this bull-run reflects in the innovations their underlying blockchain projects are developing -- that ultimately strengthens their fundamentals.

NEAR is the token of NEAR Protocol, founded by two former Google AI engineers Illia Polosukhin and Alexander Skidanov. Polosukhin was a part of Google’s ‘Transformer 8’, often regarded as one of the founding fathers of modern AI. The Transformers were the authors of the 2017 paper ‘Attention Is All You Need’ that introduced the Transformer architecture, which later became the foundation of modern AI
models like BERT and GPT.

In 2018, Polosukhin and Skidanov started NEAR Protocol, a Layer 1 blockchain over which developers can build decentralised applications (dApps). It uses sharding mechanism that makes the blockchain efficient, scalable, offering lower transaction fees, robust developer tools and requiring less computing power as the load is distributed across a network of computers.

FET is the AI token behind Fetch.ai which is eventually going to be a part of Artificial Superintelligence Alliance (ASI token), that also merges SingularityNET (AGIX), and Ocean Protocol (OCEAN). This merger is aimed at build an AI ecosystem, that will lead to the creation of Artificial General Intelligence (AGI).

In its current avatar, Fetch.ai is a decentralized network that uses AI agents to build and deploy AI-driven services in a permission-less environment. It connects people, devices, and services, across various sectors like decentralized finance (DeFi), smart energy, and transportation by enabling data sharing and AI-powered coordination.

Fetch.ai was founded in 2017 by Humayun Sheikh, Toby Simpson, and Thomas Hain. Sheikh was also a founding investor in DeepMind.

TAO is the AI coin of open source decentralised protocol Bittensor. It has built a global marketplace of machine intelligence by connecting AI model contributors (miners) and validators through a tokenized network. The platform brings together various specialized marketplaces called "subnets," where AI tasks are performed and evaluated.

Bittensor was founded in 2016 by Jacob Robert Steeves and Ala Shaabana.

ICP is the AI token of Internet Computer Protocol – a decentralised public blockchain network that lets developers build and host software and services directly on the internet. It aims to replace traditional cloud infrastructure with a blockchain-based system that will provide cloud-like functionality and performance. It was founded in 2016 by Dominic Williams.

While these are some of the mature AI tokens and their related blockchain protocols, since mid-2024, many other AI tokens were launched that are yet to prove their mettle, said Vikram Subburaj, Co-founder and CEO, Giottus.

Investors wait for the dust to settle

“Investors should not go blind on this, it’s not like everything will work,” Kashif Raza, founder of crypto education platform Bitinning said. Many companies have rebranded their existing projects as AI projects to attract funds from the VCs.

Raza added that most AI tokens are unlikely to do well in the long-run, only some will succeed.

According to Subburaj, trying to build AI models by processing data in a decentralised way will be a lot “tougher” problem to solve for compared to how OpenAI is doing it in a centralised way.

“But the benefits of decentralising AI will far surpass the efforts needed to develop it, and the crypto sector is trying to achieve that,” he said.

Mudrex’s Patel sees great potential in AI-based projects becoming a big part of the crypto ecosystem, but it won’t happen without facing some roadblocks as the tokens’ performance will be directly linked to its real-world utility this time.

“As the sector matures, projects that solve real-world problems, such as data sharing, decentralised computing, and automation, will continue to thrive. However, when looked at as an investment, the sector is cyclical just like many other sectors,” he said.

Hype-driven or purely speculative coins may see short-lived momentum as the market becomes more discerning, said Srihari.

For investors, the challenge is to spot fundamentals early and back projects creating lasting value rather than chasing the latest trend.