Review of recent popular Crypto+AI projects, these three trends have changed significantly.

Niche segmentation scenarios become the focus of expansion, with generalized AI giving way to specialized AI.

Written by: Haotian

Reviewed several popular projects in the Crypto+AI sector over the past month and found three significant trend changes, along with brief introductions and comments on the projects:

1. The project's technological path is more pragmatic, focusing on performance data rather than purely conceptual packaging;

2. Vertical segmentation scenarios become the focus of expansion, with generalized AI giving way to specialized AI;

3. Capital values business model validation more, projects with cash flow are obviously more favored;

Attachment: Project Overview, Highlight Analysis, Personal Comments:

1、 @yupp_ai

Project Introduction: A decentralized AI model evaluation platform, completed a $33 million seed round in June, led by a16z, with participation from Jeff Dean.

Highlight Analysis: Apply the subjective judgment advantage of humans to the evaluation shortcomings of AI. By using human crowdsourcing to score over 500 large models, user feedback can be redeemed for cash (1000 points = 1 dollar), which has attracted companies like OpenAI to purchase data, creating real cash flow.

Personal review: The project has a relatively clear business model and is not purely a money-burning model. However, preventing fake orders is a big challenge, and the anti-witch attack algorithm needs continuous optimization. But from the scale of the 33 million USD financing, it is clear that capital places more value on projects with monetization verification.

2、 @Gradient_HQ

Project Introduction: A decentralized AI computing network that completed a $10 million seed round in June, led by Pantera Capital and Multicoin Capital.

Highlight Analysis: With the Sentry Nodes browser plugin, there is already a certain market consensus in the Solana DePIN field. Team members come from Helium and others. The newly launched Lattica data transmission protocol and Parallax reasoning engine have made substantial explorations in edge computing and data verifiability, reducing latency by 40% and supporting heterogeneous device access.

Personal comment: The direction is right, just in line with the trend of AI localization "sinking". However, when handling complex tasks, efficiency should be compared with centralized platforms, and the stability of edge nodes is still a problem. However, edge computing is a new demand that has emerged from the web2AI competition and is also an advantage of the distributed framework of web3AI. I am optimistic about promoting landing with specific products that demonstrate actual performance.

3、 @PublicAI_

Project Introduction: A decentralized AI data infrastructure platform that incentivizes global users to contribute multi-domain data (medical, autonomous driving, voice, etc.) through token rewards, accumulating over 14 million dollars in revenue and establishing a network of over a million data contributors.

Highlight Analysis: Technically integrates ZK verification with BFT consensus algorithm to ensure data quality, and also uses Amazon Nitro Enclaves privacy computing technology to meet compliance requirements. Interestingly, it has launched the HeadCap brainwave collection device, expanding from software to hardware. The economic model is also well designed, allowing users to earn $16 + 500,000 points for 10 hours of voice annotation, while enterprises can reduce the cost of data service subscriptions by 45%.

Personal review: I feel that the greatest value of this project lies in addressing the real demand for AI data annotation, especially in fields such as healthcare and autonomous driving, which have very high requirements for data quality and compliance. However, a 20% error rate is still higher than the 10% of traditional platforms, and the fluctuation in data quality is an issue that needs to be continuously addressed. The direction of brain-computer interfaces has quite a bit of imagination, but the execution difficulty is also considerable.

4、 @sparkchainai

Project Overview: A distributed computing network on the Solana chain, completed a $10.8 million financing in June, led by OakStone Ventures.

Highlight Analysis: By aggregating idle GPU resources through dynamic sharding technology, it supports large model inference like Llama3-405B, with costs 40% lower than AWS. The design of tokenized data trading is quite interesting, as it directly transforms computing power contributors into stakeholders, which can also incentivize more people to participate in the network.

Personal comment: A typical "aggregating idle resources" model, which makes sense logically. However, a 15% cross-chain validation error rate is indeed a bit high, and the technical stability needs further refinement. Nevertheless, there are advantages in scenarios like 3D rendering where real-time requirements are not high. The key is whether the error rate can be reduced; otherwise, even the best business model will be hindered by technical issues.

5、 @olaxbt_terminal

Project Introduction: AI-driven cryptocurrency high-frequency trading platform, completed a seed round of $3.38 million in June, @ambergroup_io

Lead investment.

Highlight Analysis: The MCP technology can dynamically optimize trading paths, reduce slippage, and has demonstrated a 30% increase in efficiency. In line with the #AgentFi trend, it has found an entry point in the relatively untapped niche of DeFi quantitative trading, thereby filling a market demand.

Personal comment: The direction is correct; DeFi indeed needs smarter trading tools. However, high-frequency trading requires extremely high demands for latency and accuracy, and the real-time synergy of AI predictions and on-chain execution still needs to be verified. Additionally, MEV attacks pose a significant risk, and technical protective measures must keep pace.