Beyond the already complex challenge of implementing AI, some companies have started analyzing the possible benefits of building an AI Decentralized Autonomous Organizations (AI DAOs).
During my latest mission, I had to help create new business models, identify the right AI approach, and create a roadmap for the creation of several AI DAOs proof of concepts. Indeed, we believe that these data-driven and fully-automated organizations will become a major threat to most traditional organizations in the years to come.
在我的最新任务中，我必须帮助创建新的业务模型，确定正确的AI方法，并为创建多个AI DAO概念证明创建路线图。 的确，我们相信，在未来几年中，这些由数据驱动且完全自动化的组织将成为大多数传统组织的主要威胁。
In this article, I will help you understand what we mean by Decentralized Autonomous Organizations (DAOs), their strategic importance from a business model perspective as well as the key role of AI in the rise of DAOs.
权力下放的自治组织 (Decentralized Autonomous Organization)
Let’s start by defining the key concept of this article.
Decentralized autonomous organization (DAO): An organization that is run through rules encoded as computer programs called smart contracts. (1)
分散式自治组织 ( DAO )：通过规则编码的组织，这些规则编码为称为智能合约的计算机程序。 ( 1 )
The goal of a DAO is to create an organization that can function without “human” hierarchical management.
In theory, any interaction between humans and organizations can be expressed as a contract. Smart contracts (information transmission and enforcement of contracts) built on Blockchain technology enable us to construct these kinds of organizations based on the cloud. The goal is to be able to automate all management and administrative functions. The AI aspect of an AI DAO can be related to independent agents that autonomously take decisions.
从理论上讲，人与组织之间的任何交互都可以表示为合同。 基于区块链技术的智能合约(信息传输和合约执行)使我们能够基于云构建此类组织。 目标是能够使所有管理和管理功能自动化。 AI DAO的AI方面可以与自主进行决策的独立代理相关。
DAOs represent not only a technological revolution. Indeed, an AI DAO could create its own products and services using AI agents and sell them while the profits would go to human beings. I believe that DAOs will play a significant role as soon as universal income becomes implemented in several countries.
DAO不仅代表着技术革命。 的确，一个AI DAO可以使用AI代理创建自己的产品和服务，然后出售它们，而利润却归人类所有。 我相信，一旦在几个国家实施普遍收入，DAO将发挥重要作用。
Three ways to be involved in a DAO.
1. You can buy shares/ cryptocurrency/ tokens 2. They can be granted to you3. You can earn them performing specific tasks for the DAO
The earning part can be related to either active or passive work. For instance, finding bugs, developing software or any task that is required by the DAO. Passive working can mean sharing something, such as your computer processing cycles, storage, or even your data.
收入部分可以与主动或被动工作有关。 例如，查找错误，开发软件或DAO所需的任何任务。 被动工作可能意味着共享某些东西，例如您的计算机处理周期，存储甚至数据。
Indeed, a DAO is a computer algorithm that implements token ownership rights, contractual obligations, and business logic rules. When all these things are combined, we obtain an autonomous, data-driven, transparent company run via smart contracts that distribute value among its virtual shareholders.
Smart Contracts: A self-executing contract with the terms of the agreement between buyer and seller being directly written into lines of code. The code and the agreements contained therein exist across a distributed, decentralized blockchain network. (2)
智能合约： 一种自动执行的合约，买卖双方之间的协议条款直接写入代码行中。 其中包含的代码和协议跨分布式，分散的区块链网络存在。 ( 2 )
Based on my experience, only a few decentralized autonomous organizations (DAOs) already exist, but their rules are well established as smart contracts. The organization can perform actions, but I have not seen a system that makes independent decisions. Most of the time, it follows the rules the smart contract developer wrote.
根据我的经验，只有几个分散的自治组织(DAO)已经存在，但是它们的规则已经很好地建立为智能合约。 该组织可以执行操作，但是我还没有看到可以做出独立决策的系统。 大多数时候，它遵循智能合约开发人员编写的规则。
Our goal is to build an organization that requires no human input and can not only function well but also makes independent thoughtful changes to its structure.
战略重要性 (Strategic Importance)
For several reasons, we believe that the ability to develop a DAO will be fundamental in the future. Compared with current organizatins, DAOs have several competitive advantages.
Due to the lack of hierarchical structure, the innovation process within an DAO is potentially significantly better compared with a traditional organization. In a DAO, every innovative idea can be put forward by anyone and considered by the entire organization.
From an operational cost, having humans at the edges while benefiting from automation and AI independent agents is a major game-changer.
从运营成本上讲， 让人们处于优势地位，同时受益于自动化和独立于AI的代理，这是一个主要的游戏规则改变者 。
DAOs represent a new step in the evolution of business organizations. We believe that the convergence of several technologies (AI, Blockchain, etc.) will create not only new business models but also new types of organizations that could compete with some of our business units.
As such, it is key for large firms to start anticipating the impact of DAOs. We also expect to see a growing number of small independent DAOs in which everyone can easily invest. For this reason, it could be strategic to create our own AI DAOs and share ownership with our customers. This shift would have major repercussions on the way customers and organizations interact with each other.
因此，对于大公司而言，开始预见DAO的影响至关重要。 我们还希望看到越来越多的小型独立DAO，每个人都可以轻松投资。 因此，创建我们自己的AI DAO并与我们的客户共享所有权可能具有战略意义。 这种转变将对客户和组织之间的交互方式产生重大影响。
We might enter an era in which most companies could be run by an AI (in the future, probably an AGI) and interact with each other. This potential “AI-to-AI economy” (3) represents a major threat to our existing business models. Many companies could disappear due to a lack of competitiveness compared with DAOs. We ask ourselves questions such as:
我们可能会进入一个时代，在这个时代中，大多数公司都可以由AI(将来可能是AGI)来运作，并相互交流。 这种潜在的“ 人工智能到人工智能经济 ”( 3 )对我们现有的商业模式构成了重大威胁。 与DAO相比，许多公司可能会由于缺乏竞争力而消失。 我们问自己一些问题，例如：
- How can we remain competitive against an organization that better leverages data and has way less operational costs? 我们如何才能与组织更好地利用数据并降低运营成本的组织保持竞争力？
How would our customers react to an AI DAO selling similar products?
How would our customers react to an AI DAO selling similar products? “Why buy products from a company for which you are just a customer when you can buy these same products from an AI DAO and be an investor?”
我们的客户对销售类似产品的AI DAO有何React？ “当您可以从AI DAO购买相同产品并成为投资者时，为什么要从您只是客户的公司购买产品？ ”
- Can we build “internal” AI DAOs and transform our customers into investors? 我们可以建立“内部” AI DAO并将客户转变为投资者吗？
新业务模型，路线图和用例 (New Business Models, Roadmap and Use Cases)
Theoretically, I believe AI DAOs are the most cost-effective and open business model ever conceived. Due to the nature of DAOs (requires no employees or executive managers), these organizations can survive on almost impossible margins, and only need to cover the cost of existing.
从理论上讲，我相信AI DAO是有史以来最具成本效益和开放性的业务模型。 由于DAO的性质(不需要员工或执行经理)，这些组织可以勉强维持生存，仅需支付现有成本即可。
Any business can benefit from a model with DAO-like ambitions.
Our roadmap is to implement an AI DAO concept progressively. It could begin with the automation of a small percentage of both managerial and administrative roles, but these percentages would increase over time as the company becomes more data-driven and smart contracts handle more and more complex tasks.
我们的路线图是逐步实施AI DAO概念。 它可以从一小部分管理和行政角色的自动化开始，但是随着公司变得越来越受数据驱动和智能合约处理越来越复杂的任务，这些百分比将随着时间的推移而增加。
Moreover, we view DAOs as a construct that will have degrees of purity in its implementation. There will be cases where only a % of a company operates like one.
When it comes to concrete use cases, I have selected/identified the following ones:
Use case #1 — MarketingAn AI DAO in which where the AI selects the best companies or users to place ads with. After each marketing cycle, the AI would evaluate the ROI and adjust its marketing actions accordingly. The idea would be to create a virtuous circle, thanks to a feedback loop, to help the organization always adapt.
用例1-营销 AI DAO，AI在其中选择最佳公司或用户来投放广告。 在每个营销周期之后，AI将评估ROI并相应地调整其营销行动。 借助反馈循环，该想法将是创建一个良性循环，以帮助组织始终适应。
Use case #2 — ArtUsing generative models (GANs), we can create AI DAOs that trade their creations and distribute profits as cryptocurrency tokens to their shareholders. An AI could identify new trends (NLP on social media), create its own object(3D printing), and sell it online using autonomous agents (specific website). The profit would be distributed using cryptocurrencies.
用例2 —艺术使用生成模型(GAN)，我们可以创建AI DAO，以交易其创作并将利润作为加密货币代币分配给其股东。 AI可以识别新趋势(社交媒体上的NLP)，创建自己的对象(3D打印)，并使用自主代理(特定网站)在线出售它。 利润将使用加密货币进行分配。
Use case #3 — Vending Machine
An AI DAO related to vending machines could be used to not only takes money and deliver a snack in return but also use that money to automatically re-order the goods. The machine would also manage cleaning services and pay its rent all by itself. It has no managers, all of those processes were pre-written into code. (
与自动售货机相关的AI DAO不仅可以用来收钱和交付零食来回报，还可以使用这笔钱自动重新订购商品。 该机器还将管理清洁服务，并自行支付租金。 它没有管理器，所有这些过程都已预先编写为代码。 (
We are on the verge of a massive revolution when it comes to business models. Indeed, it would become possible for people to select and contribute to hundreds of different business models at the same time.
We envision a future in which people could search for any type of business they like, evaluate the different roles available, and/or invest in them. Compensation would be performance-based, allowing each individual to fully control the income they receive.
As you may already know, most current AI solutions help in the decision-making process but rarely learn from their actions and optimize the decisions made by themselves. Often, the most obvious way to fix this situation is to re-train the model based on newly available data and labels.
您可能已经知道，大多数当前的AI解决方案都有助于决策过程， 但很少能从其行动中学习并优化自己做出的决策。 通常，解决此问题的最明显方法是根据新近可用的数据和标签重新训练模型。
Adaptive ML and improvements in understanding causality might be required to address the capability to learn from mistakes. As such, I would not say that AI is already mature for a complex AI DAO.
为了解决从错误中学习的能力，可能需要自适应ML和对因果关系的理解上的改进。 因此，我不会说AI对于复杂的AI DAO已经很成熟。
We have built our PoCs of decentralized organizations using AI agents responsible for carrying out network decisions. This approach offers more scalability. For instance, rather than voting for every action, “investors can simply relay their preferences to the AI agent” (5). These millions of micro-decisions can then be autonomously addressed by the network’s AI agents.
我们已经使用负责执行网络决策的AI代理建立了分散组织的PoC。 这种方法提供了更大的可伸缩性。 例如，“而不是为每个动作投票”，“投资者可以简单地将其偏好传达给AI代理”( 5 )。 然后，网络的AI代理可以自动解决数百万个微决策。
Ideally, we are trying to build a system, in which different AIs could look for different parameters (e.g. branding, human resources, etc.) and take the best possible decisions both for the employees and shareholders.
Indeed, AI DAOs can combine multiple algorithms performing different sub-tasks and have access to the training data exchanged on the network. The idea is to create one gigantic feedback loop where the system would constantly learn from actions and customer data.
实际上，AI DAO可以组合执行不同子任务的多种算法，并可以访问网络上交换的训练数据。 这个想法是创建一个巨大的反馈循环 ，系统将不断从操作和客户数据中学习。
AI is also essential to the idea of self governance. A particular example is when the DAO discusses projects that are related to funding. In this scenario, the autonomous agent must first confirm that they are in harmony with the core values and goals.
人工智能对于自我治理的思想也至关重要。 一个特定的例子是DAO在讨论与资金相关的项目时。 在这种情况下，自治代理必须首先确认它们是否与核心价值观和目标保持一致。
We have identified two to three possible AI architectures related to AI. However, others can exist. Moreover, one path can be combined with other paths. (6)
我们已经确定了与AI相关的两到三种可能的AI体系结构。 但是，其他可能存在。 而且，一个路径可以与其他路径组合。 ( 6 )
The first one is impossible to achieve today since we have not succeeded yet in the creation of an Artificial General Intelligence (AGI). Imagine an AGI system leveraging smart contracts, in charge of running the organization and interacting with token holders (humans).
由于我们还没有成功创建人工智能 (AGI)，所以今天不可能实现第一个。 想象一下一个利用智能合约的AGI系统，该系统负责管理组织并与代币持有者(人类)互动。
Artificial General Intelligence: A machine capable of understanding the world as well as any human, and with the same capacity to learn how to carry out a huge range of tasks.(7)
人工智能：一种能够理解世界以及人类的机器，并且具有学习如何执行各种任务的相同能力。( 7 )
As of today, we believe that relying on AI agents seems to be the most scalable solution. Our internal projects are based on this approach. We often rely on reinforcement learning algorithms.
截止到今天，我们相信依靠AI代理似乎是最具可扩展性的解决方案。 我们的内部项目基于这种方法。 我们经常依靠强化学习算法。
AI Agents: Autonomous entity that acts, directing its activity towards achieving goals, upon an environment using observation through sensors and consequent actuators. Intelligent agents may also learn or use knowledge to achieve their goals. They may be very simple or very complex.
AI代理：一种自治实体，通过使用传感器和相应的执行器进行观察，在环境中采取行动，将其活动指向实现目标。 智能代理也可以学习或使用知识来实现其目标。 它们可能非常简单或非常复杂。
缺点与障碍 (Disadvantages & Obstacles)
While trying to build a first DAO proof of concept, I realized that there are many issues to tackle such as technology limitations, complexity and ethical issues :
First of all, it is key to remind everyone that there is still little understanding and use cases of this technology. For this reason, failures are highly possible. From a legal point of view, there is no legal classification for DAOs yet, this is an issue because even if we succeed, we would not be able to legally this autonomous entity…
首先，关键是要提醒每个人对此技术的了解和使用案例仍然很少。 因此，极有可能发生故障。 从法律的角度来看，对于DAO尚无法律分类，这是一个问题，因为即使我们成功了，我们也将无法合法地将该自治实体合法化……
As you can imagine, the development of a DAO brings many issues related to data security and domain-specific knowledge. Since all of the code would be visible and easily accessed on the blockchain, known security vulnerabilities could be exploited by hackers until all the participants, through consensus, agree to bug-fixing initiatives. We are currently trying to leverage Homomorphic encryption to fix this issue.
可以想象，DAO的开发带来了许多与数据安全性和特定领域知识相关的问题。 由于所有代码都可以在区块链上看到并轻松访问，因此黑客可以利用已知的安全漏洞，直到所有参与者通过共识同意漏洞修复计划为止。 我们目前正在尝试利用同态加密来解决此问题。
Homomorphic encryption: A cryptographic method that allows mathematical operations on data to be carried out on cipher text, instead of on the actual data itself. (8)
同态加密：一种加密方法，它允许对数据进行数学运算，而不是对实际数据本身进行密文。 ( 8 )
The manufacturing part of a DAO is still limited. I mean that an organization dealing with physical products will always require human labor until robots become cheaper and more accessible. The business case is still not obvious enough to convince internal C-level executives to massively invest.
DAO的制造部分仍然受到限制。 我的意思是，处理物理产品的组织将始终需要人工，直到机器人变得更便宜，更容易获得为止。 商业案例仍然不够明显，无法说服内部C级主管进行大规模投资。
Another challenge is the ever-increasing degrees of complexity regarding today’s organizations. Indeed, business processes are getting more complex, and so a properly self-governing DAO has much more to consider when it comes to smooth, fair operation. Our goal is to make governance mechanisms simplified and optimized.
另一个挑战是当今组织的复杂程度不断提高。 确实，业务流程变得越来越复杂，因此，要实现平稳，公平的运营，适当的自治DAO还需要考虑更多因素。 我们的目标是简化和优化治理机制。
Another key challenge while building a DAO proof of concept is to define all the rules of operating the day-to-day business. It is a very complex and tedious task. Ideally, your proof of concept can take several months before becoming a reality because it takes time to define and evaluate all these small tasks.
建立DAO概念证明时的另一个主要挑战是定义日常业务运营的所有规则。 这是一个非常复杂而乏味的任务。 理想情况下，您的概念验证可能要花费几个月才能成为现实，因为定义和评估所有这些小任务都需要时间。
In general, a new DAO is like a startup. It requires a product/ market fit and a solid business model. Another issue is related to the coordination of the different AI agents to achieve the overall system goal. Issues can arise even though the rules are clear and enforced.
通常， 新的DAO就像一家初创公司。 它需要产品/市场适应性和牢固的业务模型 。 另一个问题与不同AI代理的协调以实现整体系统目标有关。 即使规则明确并得到执行，也会出现问题。
We also question ourselves about the role of genetic algorithms that could be used for updating code. Moreover, we also want to determine if DAOs should try to maintain balances in other currencies, or should they only reward behavior by issuing their internal token?
For the above-mentioned reasons, we have a long way to go before AI DAOs become scalable business opportunities and gain traction, but the trend towards decentralized power structures increases the likelihood that these types of organizations will soon be possible.
I believe large firms should already try to build proofs of concept to better understand how this new organization can impact their business models.
有关更多信息，我建议以下链接： (For more information, I recommend the following links:)