Two Departments Jointly Issue 'Network Evaluation Activity Norms' 两部门联合印发《网络测评活动规范》

The sudden joint release of the "Regulations on Online Review Activities" by the Cyberspace Administration of China and the State Administration for Market Regulation may sound promising at first, but thinking deeper, it’s like adding fuel to the fire in the already muddled AI industry. The rules aim to maintain fair competition and protect rights, but the reality? Online reviews, particularly of AI products, have turned into a ruthless arena driven by capital and traffic. Of those "mind-blowing AI experiences" or "epic failures" you encounter, how many are authentic assessments, and how many are sponsored ads or targeted smear campaigns? The regulations are overdue but finally provide a framework. Yet, the real challenge lies in whether a piece of paper can control the review bloggers profiteering from the shadows.

Look at the absurdities in the AI industry today. A trending headline reads, "After using AI, companies seem even poorer." This is a sharp slap in the face to the current AI hype. How many business owners have been lured into launching AI projects, pouring money into servers and hiring algorithm engineers, only to see little efficiency improvement, soaring costs, and ultimately lower profits than before? AI has become a money-burning black hole for some companies—employees overtime tuning parameters, executives fretting over reports—while the so-called "smart upgrade" drags them deeper into the mire. This isn’t an isolated case; it’s a widespread bubble blown up by capital, destined to burst.

Even more ironic is the 600% surge in prices for street stall equipment, with night market vendors flooding back. What does this signify? The real economy is struggling to survive, while the AI industry continues to paint pie-in-the-sky dreams. On one side, ordinary people are hustling to make a living; on the other, tech giants are hyping concepts, poaching talent—like the chip veteran from OpenAI jumping to Anthropic. Such talent mobility may seem like industry vitality but is actually a sign of intensifying internal competition. As big corporations engage in cutthroat poaching and smaller ones struggle to survive, the AI race becomes increasingly crowded, potentially leading to a monopoly by a few giants that stifles innovation.

Returning to the "Regulations," their original intent may be good—to curb review chaos and provide a fair stage for AI products. But in reality, reviews have long been corrupted. Some bloggers hype mediocre features as revolutionary breakthroughs for money; competitors hire trolls to post malicious negative reviews, creating a toxic environment. If enforcement is weak, these rules may end up as empty talk. More critically, AI reviews require professional expertise—not just anyone can weigh in. Nowadays, any tech blogger dares to critique large language models, misleading consumers and exacerbating information asymmetry. The regulations might raise the bar slightly, but they treat symptoms, not the root cause. The industry needs greater transparency and endorsement from independent third-party institutions.

Ultimately, the AI industry resembles an athlete on steroids—running too fast, forgetting why they started. Capital chases short-term returns, companies blindly follow trends, and reviews become marketing tools, while truly impactful AI applications—like in healthcare and education—advance slowly. The trending news, "Trillion-dollar AI companies ban AI use during interviews," is a biting irony: companies use AI to screen resumes but prohibit candidates from doing the same—what a double standard! AI’s original purpose was to enhance efficiency, yet it has become a tool for some to manipulate information and harvest traffic.

In my view, rather than rushing to implement regulations, it’s better to let the industry cool down. AI is not a miracle cure; don’t expect it to change the world overnight. Companies should pause and calculate: how much real value has the investment in AI delivered? Review bloggers should uphold professional ethics and stop turning black into white for clicks. As for regulators, issuing documents isn’t enough—real crackdowns on false advertising and unfair competition are necessary. Otherwise, the "Regulations" may become another performative gesture, and the AI bubble will keep swelling until it bursts, leaving everyone in a mess.

国家网信办和市场监管总局突然联手发布《网络测评活动规范》，这事儿乍一听挺正能量，但仔细一想，简直是给现在乱成一锅粥的AI行业又撒了一把辣椒面。规范说是要维护公平竞争、保护权益，可现实呢？网络测评，尤其是AI产品的测评，早就成了资本和流量的修罗场。你刷到的那些“神级AI体验”或“翻车实录”，有多少是真实测评，有多少是充值广告或黑公关？这规范来得晚了点，但总算给了个说法，可问题在于，光靠一纸文件，能管住那些躲在屏幕后疯狂恰饭的测评博主吗？

看看现在AI行业的怪象就知道了。热榜上挂着“用了 AI 之后，公司好像更穷了”，这标题简直是对当前AI狂热的一记耳光。多少企业老板被忽悠上马AI项目，砸下真金白银买服务器、招算法工程师，结果呢？效率没见提升，成本倒是飙升，最后算算账，利润还不如以前手动操作时高。AI成了某些公司的烧钱黑洞，员工加班调参，老板对着报表发愁，而所谓的“智能升级”反而把公司拖进了泥潭。这不是个别现象，是普遍泡沫——资本吹起的泡泡，迟早要破。

更讽刺的是，地摊设备价格暴涨600%，夜市摆摊大军浩浩荡荡回归。这说明了什么？实体经济在挣扎求生，而AI行业还在云端画大饼。一边是普通人为了生计摆摊叫卖，另一边是科技巨头们炒概念、抢人才，OpenAI的芯片元老跳槽到Anthropic，这种人才流动看似行业活跃，实则是内卷加剧。大公司疯狂挖角，小公司活不下去，AI赛道越挤越窄，最后可能只剩几家巨头垄断市场，创新反而被扼杀。

回到《规范》本身，它的初衷或许是好的，想刹住测评乱象，让AI产品有个公平的展示舞台。但现实中，测评早就变质了。有些博主收钱就吹，把普通功能说成革命突破；有些竞争对手雇水军恶意差评，搞得乌烟瘴气。这规范如果执行不力，很可能又是一纸空文。更关键的是，AI测评需要专业门槛，不是谁都能插一脚的——现在随便一个数码博主就敢对大模型指指点点，结果误导消费者，加剧信息不对称。规范或许能提高一点门槛，但治标不治本，行业更需要的是透明度和第三方独立机构背书。

说到底，AI行业现在就像个打了鸡血的运动员，跑得太快，却忘了为什么出发。资本追逐短期回报，公司盲目跟风，测评沦为营销工具，而真正需要AI赋能的场景——比如医疗、教育——反而进展缓慢。热榜上“万亿美元AI公司，面试时禁止使用AI”这种新闻，更是辛辣的讽刺：公司自己都在用AI筛选简历，却禁止面试者用AI，这双标玩得溜啊。AI的初衷是提升效率，现在却成了某些人操控信息、收割流量的工具。

我倒觉得，与其急着出台规范，不如让行业先冷一冷。AI不是万能药，别指望它一夜之间改变世界。公司们该停下脚步，算算账：投入AI的钱，到底换来了多少实际价值？测评博主们也该有点职业操守，别为了流量就把黑的说成白的。至于监管，光靠发文不够，得动真格打击虚假宣传和不正当竞争。否则，《规范》最终可能变成又一场形式主义的表演，而AI泡沫继续膨胀，直到某天“砰”的一声，炸得所有人一脸狼狈。