jakiro
Author: Zhang Feng
Artificial intelligence (AI) is undoubtedly the hottest technology trend in the world. AI technology is reshaping all walks of life at an unprecedented speed.However, behind the bustle of prosperity, a cruel reality is that the vast majority of AI businesses, especially startups, have not found a stable and sustainable path to profitability.They have fallen into a dilemma of “applause but not success”, with technological prosperity and commercial losses coexisting.
1. Why “lose money and earn money”?
The profit dilemma of the AI business is not due to the failure of the technology itself, but to the structural contradictions caused by its centralized development model.Specifically, it can be attributed to the following three major reasons:
Extreme centralization: sky-high costs and oligopoly.The current mainstream AI, especially large models, is a typical “asset-heavy” industry.Its training and inference process requires huge amounts of computing power (GPU), storage and electricity.This has led to polarization: on one end are technology giants (such as Google, Microsoft, OpenAI) with abundant capital, which can afford hundreds of millions or even billions of dollars in investment; on the other end are a large number of start-up companies, which have to “tribute” most of their financing to cloud service providers to obtain computing power, and their profit margins are extremely squeezed.This model forms a “computing power oligarchy” and stifles the vitality of innovation.For example, even OpenAI relied heavily on Microsoft’s huge investment and Azure cloud computing resources in its early stages of development to support the development and operation of ChatGPT.For most players, high fixed costs make it difficult to achieve scale profitability.
Data dilemma: quality barriers and privacy risks.The fuel of AI is data.Centralized AI companies usually face two major problems in order to obtain high-quality and large-scale training data.First, data acquisition costs are high.Whether it is through paid collection, data annotation, or the use of user data, it involves a huge investment of money and time.Second, data privacy and compliance risks are huge.As global data regulations (such as GDPR and CCPA) tighten, the collection and use of data without the explicit authorization of users may trigger legal proceedings and huge fines at any time.For example, many well-known technology companies have faced sky-high fines due to data usage issues.This creates a paradox: AI cannot be developed without data, but obtaining and using data is difficult.
Imbalanced value distribution: Contributors and creators are excluded from benefits.In the current AI ecosystem, value distribution is extremely unfair.The training of AI models relies on behavioral data generated by countless users, content (text, pictures, code, etc.) produced by creators, and open source code contributed by developers around the world.However, these core contributors receive almost no return from the huge business value created by AI models.This is not only an ethical issue, but also an unsustainable business model.It dampens the enthusiasm of data contributors and content creators, and in the long run will erode the foundation for continuous optimization and innovation of AI models.A typical case is that many artists and writers have accused AI companies of using their works for training and profit without any compensation, which has triggered widespread controversy and legal disputes.
2. New paradigm of profit
DeAI (Decentralized AI) is not a single technology, but a new paradigm that integrates blockchain, cryptography and distributed computing.It aims to reconstruct the production relationship of AI through a decentralized approach, thereby solving the above three major pain points in a targeted manner and opening up the possibility of profit.
DeAI uses the “crowdsourcing” model to distribute computing power needs to idle nodes (personal computers, data centers, etc.) around the world.This is similar to “Airbnb for GPU”, forming a global and competitive computing power market that can significantly reduce computing power costs.Participants receive token incentives by contributing computing power, achieving optimal allocation of resources.
DeAI achieves “the data does not move but the model moves” through technologies such as “federated learning” and “homomorphic encryption”.Instead of centralizing raw data in one place, it distributes models to various data sources for local training, aggregating only encrypted parameter updates.This fundamentally protects data privacy while leveraging the value of decentralized data legally and compliantly.Data owners can decide independently whether to provide data and profit from it.
DeAI has built a transparent and fair value distribution system through “token economics” and “smart contracts”.Data contributors, computing power providers, model developers and even model users can automatically receive corresponding token rewards through smart contracts based on their contribution.This transforms AI from a “black box” controlled by giants to an open economy co-constructed, co-governed and shared by the community.
3. Transformation of three-tier architecture
Migrating traditional centralized AI business to the DeAI paradigm requires systematic reconstruction at the three levels of technology, business and governance.
(1) Technical reconstruction from centralized to distributed
Computing power layerRelying on the Decentralized Physical Infrastructure Network (DePIN) project, such as Akash Network, Render Network, etc., we build a flexible, low-cost distributed computing power pool to replace traditional centralized cloud services.
data layerFederated learning is used as the core training framework, combined with cryptographic technologies such as homomorphic encryption and secure multi-party computation to ensure data privacy and security.Establish a data market based on blockchain, such as Ocean Protocol, to allow data to be traded under the premise of confirmation and security.
model layerThe trained AI model is deployed on the blockchain in the form of an “AI smart contract”, making it transparent, verifiable, and callable without permission.Every use of the model and the benefits generated can be accurately recorded and distributed.
(2) Business reconstruction from selling services to ecological co-construction
From SaaS to DaaS (data as a service) and MaaS (model as a service),Enterprises no longer just sell the number of API calls, but act as ecological builders, motivating the community to participate in network construction by issuing functional tokens or governance tokens.The source of income has expanded from a single service fee to token appreciation, transaction fee dividends, etc. brought about by the growth of ecological value.
Therefore,Build a decentralized task platform to publish tasks such as data annotation, model fine-tuning, and application development for specific scenarios in the form of “bounties” for global community members to undertake and receive rewards, greatly reducing operating costs and stimulating innovation vitality.
(3) From corporate system toDAOgovernance restructuring
Based on community governance, by holding governance tokens, community participants (contributors, users) have the right to vote on key decisions, such as the adjustment direction of model parameters, the use of treasury funds, the development priority of new functions, etc.This enables true “users as owners”.
Based on openness and transparency, theAll codes, models (some can be open source), transaction records and governance decisions are put on the chain to ensure the openness and transparency of the process and establish a trustless collaborative relationship. This in itself is a powerful brand asset and trust endorsement.
Take the transformation of traditional logistics data platforms into DeAI as an example.The dilemma of the traditional logistics data platform is that although it brings together data from all parties such as shipping, land transportation, and warehousing, participants are “unwilling to share” due to concerns about the leakage of commercial secrets, resulting in data islands and limited platform value.The core of the transformation to DeAI is to release the value of data and provide fair incentives without exposing the original data:
Technically build a trusted computing network.The platform no longer centrally stores data, but transforms into a blockchain-based coordination layer.Adopting technical models such as federated learning, the AI model is “airborne” to the local servers of each enterprise (such as shipping companies and warehouses) for training, and only the encrypted parameter updates are aggregated to jointly optimize the global prediction model (such as cargo ship arrival time, warehouse liquidation risk), achieving “data does not move, value moves”.
Promote data assetization and token incentives in business.The platform is issued with practical points, and logistics companies can “mine” by contributing data (model parameters) to obtain point rewards.Downstream customers (such as cargo owners) pay tokens to query high-precision “forecast results” (for example: the on-time performance of a certain route in the next week) instead of purchasing raw data.Revenues are automatically distributed to data contributors through smart contracts.
Building industries based on governanceDAO,Key decisions (such as new feature development, fee adjustment) are jointly voted by token holders (i.e. core participants) to transform the platform from a private company-dominated to an industrial community.
The platform has transformed from a centralized organization trying to extract data intermediary fees to a nervous system for co-construction, co-governance and sharing of the entire logistics industry chain. By solving the trust problem, it has greatly improved the industry’s collaborative efficiency and risk resistance.
4. Compliance and Security
Although DeAI has great prospects, its development is still in its early stages and faces a series of challenges that cannot be ignored.
Compliance and legal uncertainty.In terms of data regulations,Even if the data does not move, models such as federated learning still need to strictly comply with the requirements of “purpose limitation”, “data minimization” and user rights (such as the right to be forgotten) in regulations such as GDPR when processing personal data.Project parties must design compliant data authorization and exit mechanisms.
In terms of securities regulations,Tokens issued by projects can easily be recognized as securities by regulatory agencies in various countries (such as the U.S. SEC), thus facing strict regulatory scrutiny.How to avoid legal risks when designing a token economic model is the key to the project’s survival.
In terms of content responsibility,If a DeAI model deployed on the chain produces harmful, biased or illegal content, who is responsible?Is it a model developer, a computing power provider or a governance token holder?This has brought new issues to the existing legal system.
In terms of security and performance challenges,model safetyThat isModels deployed on public chains may face new attack vectors, such as exploiting vulnerabilities in smart contracts or maliciously damaging federated learning systems by poisoning data.
The performance bottleneck isThe transaction speed (TPS) and storage limitations of the blockchain itself may not support high-frequency, low-latency large model inference requests.This requires an effective combination of Layer 2 expansion solutions and off-chain computing.
Collaboration efficiency isAlthough distributed collaboration is fair, decision-making and execution efficiency may be lower than that of centralized companies.How to strike a balance between efficiency and fairness is an art that needs to be continuously explored in DAO governance.
As a revolution in production relations, DeAI is expected to break the monopoly of giants through distributed technology, token economy and community governance, release idle computing power and data value around the world, and build a new AI ecosystem that is fairer, sustainable and potentially more profitable.
5. Current exploration direction
The current development of AI tools is still a long way from realizing ideal decentralized artificial intelligence.We are still in the early stages of being dominated by centralized services, but some explorations have pointed out the future direction..
Current explorations and future challenges.Although the ideal DeAI has not yet been realized, the industry is already making valuable attempts, which helps us see the future path and the obstacles that need to be overcome.
Such as the prototype of collaboration in a multi-agent system.Some projects are exploring the construction of an environment where AI agents collaborate and co-evolve.For example, the AMMO project aims to create a “symbiotic network of humans and AI.” The multi-agent framework and RL Gyms simulation environment it designed allow AI agents to learn to collaborate and compete in complex scenarios.This can be seen as an attempt to build the underlying interaction rules of the DeAI world.
Another example is a preliminary attempt at an incentive model.In the vision of DeAI, users who contribute data and nodes that provide computing power should receive fair returns.Some projects are trying to redistribute value directly to ecosystem contributors through crypto-based incentive systems.Of course, how this economic model can operate on a large scale, stably and fairly remains a huge challenge.
Another example is moving towards a more autonomousAI: Deep Research products demonstrate the powerful autonomy of AI in specific tasks (such as information retrieval and analysis).They can autonomously plan, perform multi-step operations and iteratively optimize the results. This task automation capability is the basis for the independent work of AI agents in future DeAI networks.
For AI practitioners struggling in the red ocean, instead of being entangled in the old paradigm, it is better to bravely embrace the new blue ocean of DeAI.This is not only a transformation of technical routes, but also a reshaping of business philosophy – from “extraction” to “incentives”, from “closed” to “open”, from “monopoly profits” to “inclusive growth”.
