Combining Top-Level Design & Crossing the River by Feeling the Stones - Discussing the Future of China's AI Governance - Li Qiang Calls for Supporting High-Tech Skilled Talents - Data on Energy Mix

Hi folks,

I hope you’ve all had a good weekend. I took a bit of a break late last week, celebrating Diwali. There were lots of colourful lights and much-needed sleep to enjoy.

The wonderful thing about living in a multicultural, multi-faith society is that you have so many new year’s within one year that you can always re-start and renew resolutions that didn’t work out.

So, here’s wishing everyone reading this newsletter a happy and prosperous year ahead.

Warmly,

Manoj

---

Before we get to today’s edition, I want to cover a piece from Saturday. This one deals with Xi’s engagement with officials at the provincial and ministerial level. It makes some really interesting points about the principles and boundaries of reform:

The combination of top-level design and ‘crossing the river by feeling the stones’ is one of the distinctive characteristics of comprehensively deepening reform in the new era. On October 29, 100 days after the Third Plenary Session of the 20th CPC Central Committee, at the opening session of a special seminar for provincial and ministerial-level leaders to study and implement the spirit of the Third Plenary Session, General Secretary Xi Jinping’s remarks on ‘crossing the river by feeling the stones’ were thought-provoking. ‘Some people say, at this point in time, we’re still “crossing the river by feeling the stones”. Isn’t that conservative and timid? ‘Reforms that follow familiar paths are like swimming in a pool, where lines mark shallow and deep ends clearly. But in a turbulent river, don’t we need to feel for stones to cross? It’s not that, after decades of reform and opening up, we’ve already figured out every issue. If everything were clear, why would there still be a need for reform?’ These words sound familiar. During his first local inspection after the 18th Party Congress in Guangdong, the General Secretary made it clear: ‘We cannot say that feeling the stones to cross the river was necessary in the early stages of reform and opening up, but now we can no longer mention feeling the stones to cross the river’. There are both key approaches for overall planning. In the new era of comprehensively deepening reform, an important reform methodology, which is also a major theoretical achievement we have made, is ‘combining top-level design with crossing the river by feeling the stones.’ The combination of the two reflects the important principle of governance of seeking progress while maintaining stability and has multiple practical implications.” 顶层设计和摸着石头过河结合，是新时代全面深化改革的鲜明特征之一。在党的二十届三中全会百天之后，10月29日，省部级主要领导干部学习贯彻党的二十届三中全会精神专题研讨班开班式上，习近平总书记一段“摸着石头过河”的论述发人深思。“有些人讲，都什么年月了，还搞摸着石头过河，不是保守、胆小吗？”“熟门熟路的改革，就像在游泳池，标着线，哪儿是浅水区、哪儿是深水区一目了然。到湍急的河里，不得摸着石头过河？并不是说改革开放搞了几十年了，现在什么问题都摸清楚了。如果都摸清楚了还要什么改革呢？”这番话似曾相识。党的十八大后首次地方考察到了广东，总书记就明确：“不能说改革开放初期要摸着石头过河，现在再摸着石头过河就不能提了”。既有关键处落子，也有全局上谋划。新时代全面深化改革，一个重要的改革方法论，也是我们取得的一个重大理论成果，就是“把顶层设计和摸着石头过河结合起来”。二者结合起来，体现的正是稳中求进工作总基调这一治国理政重要原则，有着多重现实指向。

Later the piece adds: “Top-level design emphasises systematic thinking and overall strategy, while ‘crossing the river by feeling the stones’ emphasises daring to explore and experimentation at the grassroots level. ‘We must combine top-level design with crossing the river by feeling the stones’ - this is an inherent requirement for further comprehensively deepening reform, and it is also an inevitable choice to enhance the systematic, holistic, and coordinated nature of reform, and to improve the cohesion, driving force, and implementation of reform.” 顶层设计更重系统思维、整体战略，摸着石头过河更重敢闯敢试、基层探索。“必须把顶层设计和摸着石头过河结合起来”，这是进一步全面深化改革的题中应有之义，也是增强改革的系统性、整体性、协同性，提升改革的凝聚力、推动力、落实力的必然选择.

The piece then stresses that during his remarks, Xi spoke about “两手并用” using both hands, adopting a dual strategy or doing two things in tandem. It adds:

“The General Secretary profoundly pointed out: ‘For things we are confident about and have thought through carefully, we should implement top-level design; for things we are uncertain about and that are not yet mature, we should cross the river by feeling the stones - these two approaches should be used in tandem’. Reform is a journey of exploration full of unknowns. The coordination of ‘using both hands’ contains the harmonisation of ‘knowing’ and ‘doing’. For what is clearly understood, we must ‘commit to advancing resolutely’. For what is not so clearly understood yet still requires breakthroughs, we should neither stand still nor rush forward blindly. Over these years, a series of major reforms have formed replicable experiences through pilot exploration, which have then evolved into institutional achievements, and finally expanded to larger scales. Comprehensively deepening reform is thus about ‘accumulating small victories into big victories’.”

Another phrase that the piece talks about is ‘establish first, then break’ 先立后破. 

Reform is a process of breaking the old and establishing the new. At this point in history, the prioritisation between breaking and establishing concerns not only the strategic planning of top-level design but also the orderly steps of crossing the river by feeling the stones. The General Secretary has a clear view: ‘To further comprehensively deepen reform, we must adhere to institutional building as the main line and emphasise the simultaneous approach of establishing and breaking, prioritising establishment before breaking. All reform measures must be fully demonstrated and carefully designed, with in-depth risk assessments, and proper control of timing, extent, and effectiveness. What needs to be established should be proactively established, and it must be established firmly, sustainably, feasibly, and effectively; what needs to be broken should be broken in a timely manner, decisively, and thoroughly on the basis of what has been established, achieving steady and swift reform progress through the unity of breaking and establishing’. 

开班式上，习近平总书记提到了两个词。一个词是“两手并用”。实践、认识、再实践、再认识的过程，也是量变到质变的过程。总书记深刻指出：“对于胸有成竹的、思虑缜密的，那就要搞顶层设计；对于心里没底的、还不成熟的，就摸着石头过河，这两件事要‘两手并用’。”改革是一场充满未知的求索。“两手并用”的统筹，蕴含着对“知”和“行”的协调。看准了的，“要下决心推进”。“看得还不那么准、又必须取得突破”的呢？既不能裹足不前，又不能盲目冒进。这些年，一系列重大改革通过试点探索形成了可复制经验，进而上升为制度性成果，最终大范围铺开。全面深化改革，就是这样“积小胜为大胜”。一个词是“先立后破”。改革是一个破旧立新的过程。历史行进到此时，关于破与立的优先序，既事关顶层设计的运筹摆布，也牵涉摸着石头过河的步骤章法。总书记看得透彻：“进一步全面深化改革，要坚持以制度建设为主线，更应突出破立并举、先立后破。各项改革举措都要充分论证、精心设计，深入开展风险评估，把握好时度效。该立的积极主动立起来，而且要立得稳、立得住，行得通、真管用；该破的在立的基础上及时破、坚决破、彻底破，在破立统一中实现改革蹄疾步稳.”

A couple of paragraphs towards the end are also interesting, since they set the boundaries for the reform agenda:

“What should not be changed? The General Secretary pointed out: ‘Our fundamental political system, basic political system, and supporting systems are all very mature now. The People's Republic of China has been established for 75 years, the People's Congress for 70 years, and the Political Consultative Conference for 75 years - do they still need major reforms? Where would further reforms lead? We shouldn’t reform just for the sake of reform, nor should we reform blindly’. What must not change? Take economic system reform for example. For example: on the issue of ‘letting the market play a decisive role in resource allocation while better leveraging the role of government’, General Secretary Xi Jinping recalled the lengthy discussion process during the preparation for the Third Plenary Session of the 18th CPC Central Committee, emphasising that ‘this important assertion is absolutely crucial.’ Another example is the consistent adherence to and implementation of the ‘Two Unwaverings’. The General Secretary also responded to the concerns of those who try to distinguish the importance of the two: ‘There’s no singling out, no one-sidedness; this has never wavered. We must be fair and just, maintain the balance, and not weigh one as more important than the other’. These things that cannot be casually changed or easily altered were also established through repeated experiences of exploration and were ultimately solidified through top-level design. The General Secretary made this assessment: ‘As the reform has progressed to this day, we already have a relatively clear big picture in terms of the top-level design of the reform’.” 什么是不该改的？总书记指出：“我们的根本政治制度、基本政治制度以及配套制度都很成熟了。新中国成立75年，人大70年、政协75年，还需要大改吗？再改改到哪去？不能为改而改，不能稀里糊涂地改。”什么是不能变的？就拿经济体制改革来说，比如，“使市场在资源配置中起决定性作用和更好发挥政府作用”，习近平总书记回忆了在党的十八届三中全会准备工作中，这句话历经长时间讨论的出炉过程，强调“这一重要论断是非常关键的”。再比如，坚持和落实“两个毫不动摇”，一以贯之。总书记也回应了那些试图分出二者轻重的关切：“没有单打一，没有一点论，这一点什么时候都没有动摇过。公平公正，一碗水端平，不要搞孰轻孰重。”这些不能动辄改、不能轻易变的，也都是历经反复摸石头，最终以顶层设计的方式夯实的。总书记作出判断：“改革推进到今天，在改革顶层设计方面，我们已经有了比较清晰的大盘子.” 

Here are the key reports and articles that I found noteworthy from the People’s Daily’s edition on Monday, November 04, 2024.

Page 1: There isn’t a lot useful on the front page today. There’s a report (English report) on Li Qiang’s visit to Shanghai, where he discussed vocational education and skilled talent training. Li visited the Shanghai Nanhu Vocational & Technical College. He said there “it is necessary to deepen the integration of industry and education as well as the cooperation between colleges and enterprises, optimize the setting of college disciplines and majors based on the needs of the society and industries’ actual development and strengthen hands-on skill practice, adding that more skilled teachers are needed to innovate the models and methods of teaching.”

Li added: 

“It is necessary to grasp the trend of the new round of scientific and technological revolution and industrial transformation, aim at the forefront of science and technology and the world’s first-class, cultivate more ‘high-tech, cutting-edge and scarce’ skilled talents with strong theoretical foundations and practical capabilities to help achieve high-level scientific and technological self-reliance and self-strengthening.” 要把握新一轮科技革命和产业变革趋势，瞄准科技前沿和国际一流，培养更多理论素养高、实践能力强的“高精尖缺”技能人才，助力实现高水平科技自立自强.

Noting that there are ever-growing demands for elderly care, childcare and medical care, Li urged efforts to strengthen the training of skilled workers in the related fields. In addition, while visiting the WorldSkills Museum, Li said that “governments at all levels should continue to refine support policies for the cultivation of skilled workers by strengthening institutional innovation and investment. Efforts should also be made to improve the systems of development planning and services for skilled workers with a view to broadening their development channels and enhancing their incomes”.

Also on the page is a long article previewing the 7th China International Import Expo. The article calls it a “major economic and diplomatic” event. One key point in the piece is:

“Opening up is the inevitable path to world prosperity and development. Looking back at history, openness and inclusiveness lead to progress, while isolation and narrowness lead to obstacles. After the Cold War, economic globalisation developed rapidly, greatly promoting the flow of goods and capital, providing strong momentum for world economic development. In today’s world of ‘you are in me, and I am in you’, opening up and cooperation is the only choice. China’s hosting of the Import Expo is a major initiative to proactively open its market to the world, and a practical action to promote the building of an open world economy and advocate inclusive economic globalisation. Only by gathering the power of cooperation through openness can we achieve great things, accomplish good things, and sustain long-term success, bringing a bright future for global development. Opening up is a distinctive marker of Chinese-style modernization. Major trading partners with more than 150 countries and regions, ranking first in total goods trade for 7 consecutive years, outward direct investment flow ranking among world’s top three for 12 consecutive years…Since reform and opening up, China has forged a path of seeking common development through openness. The Import Expo is a vivid example that ‘the more China develops, the more it opens up’, witnessing China's door of opening up growing ever wider. Only an open China can become a modernised China…” 开放，是世界繁荣发展的必由之路。回望历史，开放包容则路达，封闭狭隘则途阻。冷战结束后，经济全球化迅猛发展，极大促进了商品和资本流动，为世界经济发展提供了强劲动力。在“你中有我、我中有你”的当今世界，开放合作是唯一选择。中国举办进博会，是主动向世界开放市场的重大举措，是推动建设开放型世界经济、倡导普惠包容的经济全球化的实际行动。以开放汇聚合作之力，才能办成大事、办成好事、办成长久之事，为全球发展带来光明前程。开放，是中国式现代化的鲜明标识。150多个国家和地区的主要贸易伙伴，货物贸易总额连续7年居世界第一，对外直接投资流量连续12年居全球前三……改革开放以来，中国走出了一条在开放中谋求共同发展的道路。进博会是“中国越发展，就越开放”的生动例证，见证中国开放的大门越开越大。只有开放的中国，才会成为现代化的中国.

Page 2: There’s a report discussing China’s energy mix. Some key data points:

“In the first eight months, the investment in key energy projects was about 1.7 trillion yuan, a year-on-year increase of 16.2%. Among these, investment in energy network infrastructure has been steadily released. Investment in 5 under-construction ultra-high voltage DC transmission projects across provinces and regions is accelerating, LNG receiving terminal investments in Liaoning, Zhejiang, Guangxi, and Shandong are growing rapidly, and under-construction projects are quickly materializing into physical work progress. Non-fossil energy power generation investment is growing steadily. A batch of centralized wind power projects in Hebei, Liaoning, Heilongjiang, Gansu, and Xinjiang are accelerating, while five newly approved nuclear power projects in Shandong Zhaoyuan, Jiangsu Xuwei, Zhejiang Sanao, Guangdong Lufeng, and Guangxi Bailong are gradually materializing into physical work progress. Investment in new energy business forms is expanding in multiple areas. In the first 8 months, completed investment in new energy storage increased by 21% year-on-year, while pumped storage completed investment increased by 45.5% year-on-year. A batch of multi-energy complementary clean energy base projects in Gansu, Qinghai, and Xinjiang are progressing in an orderly manner, while Inner Mongolia's under-construction green hydrogen production projects and Henan’s under-construction hydrogen facilities projects continue to form effective investment. The completed investment in key energy projects has maintained relatively rapid growth, further consolidating the foundation of energy supply security and injecting momentum into high-quality economic and social development. As of the end of September, the national total power generation installed capacity reached 3.16 billion kilowatts, a year-on-year increase of 14.1%. Coal, oil, and gas production capacity has steadily improved, with crude oil and natural gas production increasing by 2% and 6.6% year-on-year in the first three quarters of this year. 其中，能源网络基础设施投资稳步释放。5项在建跨省跨区特高压直流输电工程投资加快推进，辽宁、浙江、广西、山东液化天然气接收站投资快速增长，在建项目加快形成实物工作量。非化石能源发电投资稳步增长。河北、辽宁、黑龙江、甘肃、新疆一批集中式风电项目加快推进，山东招远、江苏徐圩、浙江三澳、广东陆丰、广西白龙5个新核准核电项目逐步形成实物工作量。能源新业态投资多点开花。前8个月，新型储能完成投资额同比增长21%，抽水蓄能完成投资额同比增长45.5%。甘肃、青海、新疆一批多能互补清洁能源基地项目有序推进，内蒙古在建绿氢制备项目、河南在建加氢设施项目持续形成有效投资。能源重点项目完成投资额保持较快增长，进一步夯实了能源供给保障基础，为经济社会高质量发展注入动能。截至9月底，全国发电总装机容量达到31.6亿千瓦，同比增长14.1%。煤炭、油气产能稳步提升，今年前三季度，原油、天然气产量同比增长2%、6.6%.

“In the first three quarters, China’s new installed capacity of renewable energy power generation was 210 million kilowatts, a year-on-year increase of 21%, accounting for 86% of the newly-installed capacity of electricity. Among them, hydropower added 7.97 million kilowatts, wind power added 39.12 million kilowatts, solar power generation added 161 million kilowatts, biomass power generation added 1.37 million kilowatts, and the total increase in wind and solar power generation exceeded 200 million kilowatts. By the end of September, China’s renewable energy installed capacity reached 1.73 billion kilowatts, a year-on-year increase of 25%, accounting for about 54.7% of the total installed capacity.

Looking at the power generation, in the first three quarters, the national renewable energy power generation reached 2.51 trillion kilowatt-hours, an increase of 20.9% year-on-year, accounting for approximately 35.5% of the total power generation. Among them, the combined power generation from wind and solar reached 1,349 billion kilowatt-hours, a year-on-year increase of 26.3%, which is roughly on par with the electricity consumption of the tertiary industry during the same period, exceeding the electricity consumption of urban and rural residents during the same time.” 今年以来，我国能源绿色低碳转型不断取得新突破。前三季度，全国可再生能源发电新增装机2.1亿千瓦，同比增长21%，占电力新增装机的86%。其中，水电新增797万千瓦，风电新增3912万千瓦，太阳能发电新增1.61亿千瓦，生物质发电新增137万千瓦，风电太阳能发电合计新增突破2亿千瓦。截至9月底，全国可再生能源装机达到17.3亿千瓦，同比增长25%，约占我国总装机的54.7%。再看发电量，前三季度，全国可再生能源发电量达2.51万亿千瓦时，同比增加20.9%，约占全部发电量的35.5%。其中，风电太阳能发电量合计达13490亿千瓦时，同比增长26.3%，与同期第三产业用电量基本持平，超过了同期城乡居民生活用电量.

Page 9: The entire page is dedicated to views from academics as part of a roundtable discussing the establishment of “oversight systems to ensure the safety of AI” and improving the mechanisms for developing and managing generative AI, which were goals outlined in the resolution issued after the Third Plenum.

First, Liu Yanhong from the China University of Political Science and Law argues that the development of generative AI can be classified into three stages:

• the preparatory stage involving pre-training and algorithm upgrades aided by human annotation

• the operational stage where input data is processed by algorithms to produce generated outputs, and 

• the generative stage where the outputs are applied in society.

Liu says that “We must understand the operational mechanisms of generative AI in depth, grasp the characteristics of the formation and development of security risks at each stage, and use legal means to strengthen systematic governance, ensuring that the immense power contained in generative AI always operates within the framework of the rule of law.” 生成式人工智能在技术运行上可分为三个阶段，即前置性学习训练及人工标注辅助算法升级的准备阶段，输入数据进行算法处理得出生成物的运算阶段，生成物进入社会加以运用的生成阶段。我们要深入分析生成式人工智能的运行机理，把握各阶段安全风险形成与发展的特征，运用法治手段加强系统性治理，确保生成式人工智能所蕴含的巨大力量始终在法治轨道上发挥作用.

She then outlines the key risks for each stage. 

For the first stage, Liu argues that the risks are primarily related to data security. At this stage, it is important to ensure “appropriate classification of data and the establishment of usage patterns and protection methods for different types of data to properly address relevant data security risks, avoiding illegal use or improper disclosure of data, which could lead to infringement disputes.” 生成式人工智能通过数据训练、数据处理分析来提炼信息、预测趋势。这就必须对数据进行适当分类，确立不同类型数据的利用模式和保护方式，以妥善应对相关数据安全风险，避免数据违规利用或者不当泄露，产生侵权方面的纠纷.

Liu adds: “Legal rules for generative AI's acquisition and use of government data should be clarified to both meet society’s needs for utilizing government data and strongly support the development, training, and application of AI government service large models to improve public service and social governance intelligence levels, while also regulating its processing methods to avoid results derived from government data harming individual rights or disrupting social public order.” 应当明确生成式人工智能获取和利用政务数据的法律规则，既满足利用政务数据服务社会的需求，有力支持人工智能政务服务大模型开发、训练和应用，提高公共服务和社会治理智能化水平；又规范其加工方式，避免利用政务数据得出的成果侵害个人权益、破坏社会公共秩序.

On personal data collection, She argues that “generative AI tends to over-collect personal data to improve conclusion accuracy, such as analyzing medical health data to mine personal whereabouts and predict personal life trajectories. Therefore, collection must adhere to law, collecting personal data within the minimum scope needed for technology, and setting reasonable data processing depths to avoid excessive mining of potential information.” 实践中，生成式人工智能倾向于过度收集个人数据以提升结论准确性，比如，通过分析医疗健康数据来挖掘个人行踪、预测个人生活轨迹。为此，必须坚持依法收集，按照技术所需的最小范围收集个人数据，设置合理的数据处理深度，避免过度挖掘潜在信息.

For the second stage, Liu highlights the “inherent risk of algorithmic bias.” Liu says:

“To prevent and resolve algorithm bias risks, targeted governance should be conducted according to the principles and domains of algorithm bias formation. Legal regulatory requirements should be deeply embedded into the algorithmic models of generative AI, promoting technology for good, eliminating algorithmic bias, and ensuring reasonable use of generative AI algorithms and allocation of computing resources. Based on the concept of combining technology and management, strengthen full-cycle safety supervision of algorithms, implementing legal regulatory requirements throughout the entire process of generative AI operation. When setting algorithms, it is essential to adhere to relevant legal rules and technical standards, implement the regulatory requirements of ‘machine learning + human annotation’, review the algorithm modules with risks, and better discover the technical risks within the generative AI algorithm models; when congenital algorithmic biases are discovered, corrections should be made from within the algorithm based on legal requirements, ensuring that the modified algorithm can operate normally; if problems occur afterwards, conduct source-tracing governance of AI algorithms, achieve precise attribution of responsibility and make corrections, promote the improvement of regulatory standards for generative AI algorithms, fill gaps in preliminary preventive review, and use both technical and legal means in parallel to equally emphasise development and management.” 防范化解算法偏见风险，应根据算法偏见的产生原理与产生场域进行针对性治理。要将法律规范的要求深度嵌入生成式人工智能的算法模型之中，推动技术向善，消除算法偏见，确保合理利用生成式人工智能算法并分配算力资源。基于技管结合理念，加强对算法的全周期安全监管，将法律规范的要求落实到生成式人工智能运行的全流程之中。在设置算法之初就要遵循相关法律规则与技术标准，落实“机器学习+人工标注”的规范要求，审查存在风险的算法模块，更好发现生成式人工智能算法模型中的技术风险；当发现先天性算法偏见时，依据法律要求从生成式人工智能的算法内部进行纠正，确保修改后的算法能正常运行；事后出现问题时，对人工智能算法进行溯源治理，实现精准归责并加以纠正，推动完善生成式人工智能的算法监管标准，填补事前预防审查的不足，以技术手段与法律手段并行做到发展与管理并重.

For the third stage, Liu writes that there are many risks from the risk of abuse to IPR risks, exacerbating information asymmetry, widening the digital divide, and damaging the interests of digitally disadvantaged groups. In terms of IPR risks, Liu argues that “substantial analyses of the technical models and principles of generative AI should be conducted according to the standards of intellectual property law. If human intervention is necessary to ensure that the generated outputs possess originality and innovation, intellectual property rights should be granted and ownership clearly defined, reinforcing the systematic protection of intellectual property rights in the field of generative AI. At the same time, it is necessary to reasonably determine the scope of protection for the intellectual property rights of generated outputs to avoid an unlimited expansion of the protection scope, which could hinder the promotion and application of generative AI and technological development.” 为及时化解相关问题，应对生成式人工智能的技术模式、技术原理按照知识产权法的标准开展实质分析，如果技术上需要人类意志介入，使生成物能够产生独创性与创新性，应赋予知识产权并明确其归属，强化生成式人工智能领域知识产权的系统性保护；同时要合理确定对生成物知识产权保护的范围，避免保护范围无限扩张，妨碍生成式人工智能的推广运用和技术发展.

Next, Gu Liping from the School of Journalism and Communication at Nanjing Normal University discusses the importance of strengthening privacy protection in the era of artificial intelligence and ensuring personal information security. Gu argues that “the right to privacy faces severe challenges in the era of AI.” Gu elaborates the types of harms and challenges in the first bit before saying that:

In terms of solutions, Gu stresses the “principle of informed consent”. He adds:

“The principle of informed consent includes two aspects: knowledge and consent. Consent must be based on knowledge. Without sufficient knowledge and understanding, there can be no real consent. Information, understanding and voluntariness are the three elements of the principle of informed consent. On the basis of comprehensive ‘information’, individuals can independently express their ‘consent’. This requires that when users use artificial intelligence, easy-to-understand and clear prompts and instructions should be made to obtain users’ consent to the collection and use of personal information. If personal information will flow between different platforms, users need to be informed of the scope, target and use boundaries of the flow. For a good and smooth user experience, users can also be given the option of one-time or phased authorization. Users should be informed about the scope, methods, and purposes of personal information collection, as well as who their personal information will be shared with, and they should have the option to withdraw at any time. When conducting personal information analysis, users should be alerted via pop-ups or other forms and provide real-time authorization. Establishing a data lifecycle and timely deletion of personal information is also an effective way to protect the security of personal information.” 知情同意原则包括知情和同意两方面，同意必须以知情为前提，没有充分的知情和理解，就不可能有真正意义上的同意。信息、理解和自愿，是知情同意原则的三要素。在全面“知情”的基础上，个人可以自主作出怎样“同意”的意思表示。这就需要在用户使用人工智能时，作出通俗易懂而又清晰明了的提示说明，征得用户对个人信息收集使用的同意。如果个人信息会在不同平台之间流动，需要将流动范围、目标、使用边界让用户知晓。为了良好和流畅的用户体验，也可以给用户提供一次性或分阶段进行授权的选择。要告知用户收集个人信息的范围、方式和用途以及与谁共享个人信息，用户也应当可以选择随时退出。在进行个人信息分析时，应以弹窗或其他形式提示用户注意并实时授权。设置数据生命周期、按时删除个人信息也是保护个人信息安全的有效方式.

The author also calls for “protect privacy rights and personal information security by improving database security, core data encryption, personal data anonymization and other technologies”. 因此，技术解决方案必须置于关键位置，应通过完善数据库安全、核心数据加密、个人数据脱敏等技术，建立保护隐私权和个人信息安全的防火墙.

The third and final write-up is by Professor Zhang Xin from the School of Law at the University of International Business and Economics. Zhang argues that for a variety of reasons, AI entities can cause “complex, dynamic and unforeseen risks.” Therefore, she calls for a “modular governance framework” for AI. “Such a framework must cover the entire industrial chain and various levels of endpoints, starting from key nodes such as data modules, algorithm modules, and model architectures, to design corresponding governance modules.” 因此，可以构建一种能够覆盖整个产业链和各个端层的模块化治理框架，从数据模块、算法模块、模型架构等关键节点出发，设计相应的治理模块.

Zhang later adds that “AI entities are deeply interactive, highly interconnected, and dynamically adaptable. Correspondingly, the governance approach should go beyond traditional individual-centered governance and promote the formation of a governance ecosystem characterized by extensive interconnectivity, multi-party participation, and multi-level collaboration. In this context, technical communities such as developers and operations and maintenance personnel will play a crucial ‘whistleblower’ role in the governance of artificial intelligence entities. We should better leverage the supervisory advantages of technical communities and establish effective constraint mechanisms within artificial intelligence companies. Additionally, we should actively enhance the digital literacy of a wide range of users, raising their awareness of the lawful, safe, and responsible use of artificial intelligence entities, achieving positive interactions with these entities, and fostering an upward and virtuous operational state.” 人工智能体具有深度交互性、高度互联性以及动态适应性。相应地，治理方式应当超越传统的以个体为中心的治理，推动形成广泛互联、多方参与、多层次协作的治理生态。其中，技术开发人员、运营维护人员等技术社群对于人工智能体的治理将起到至关重要的“吹哨人”作用。应更好发挥技术社群的监督优势，在人工智能企业内部构建有效的约束机制。还应积极提高广大用户的数字素养，增强其依法、安全、负责任使用人工智能体的意识，实现与人工智能体的良性交互，推动形成向上向善的运行状态.