内容来源：https://techcrunch.com/2026/01/05/microsofts-nadella-wants-us-to-stop-thinking-of-ai-as-slop/

内容总结：

在韦氏词典将“AI劣质内容”评为年度词汇数周后，微软首席执行官萨提亚·纳德拉近日展望了2026年人工智能的发展方向。他在个人博客中以一贯的思辨风格提出，人们应停止将AI视作“劣质内容生成器”，而应将其看作“思维的自行车”——即人类潜能的“脚手架”而非替代品。

纳德拉呼吁业界超越“劣质与精密”的二元争论，建立新的认知平衡，将AI定位为人类协作的认知增强工具。然而现实矛盾在于：AI产品的营销往往以替代人力作为定价依据，而包括Anthropic首席执行官达里奥·阿莫代伊在内的行业领袖多次警告，AI可能导致初级白领岗位大规模流失，未来五年失业率或升至10%-20%。

当前研究数据呈现复杂图景。麻省理工学院“冰山项目”显示，目前AI仅能承担约11.7%的有偿劳动任务，且多集中于护理文书、基础编程等辅助环节。值得注意的是，先锋集团2026年经济预测报告发现，受AI影响最大的百个职业在就业增长和实际工资涨幅方面反而领先劳动力市场。该报告指出，精通AI工具者正创造更高价值。

尽管微软2025年在创收背景下裁员超1.5万人，并将“AI转型”列为核心战略，但先锋集团分析认为，裁员主因仍是企业常规战略调整——即收缩衰退业务、扩张增长领域。研究机构数据显示，2025年美国AI相关裁员约5.5万人，涉及亚马逊、Salesforce等科技企业。

值得玩味的是，社交网络上广受欢迎的AI生成短视频和表情包，恰恰印证了所谓“劣质内容”亦具备其娱乐价值。在迈向2026年之际，这场关于AI本质的讨论，仍在“替代威胁”与“增强潜能”的辩证中持续演进。

中文翻译：

在韦氏词典将"slop"（粗制滥造的内容）评为年度词汇几周后，微软首席执行官萨提亚·纳德拉对2026年人工智能的发展前景发表了看法。

纳德拉以其标志性的知性风格在个人博客中写道，他希望人们不再将AI视为"粗制滥造的内容"，而应开始将其看作"思维的自行车"。他写道："我们需要发展'思维自行车'这一概念，始终将AI视为人类潜能的脚手架，而非替代品。"他进一步阐述："我们必须超越'粗制滥造'与'精密复杂'的二元争论，建立关于'心智理论'的新平衡——这种理论应承认人类配备了这些新型认知增强工具，并以此为基础构建人际关系。"

仔细解读这些表述可以发现，纳德拉不仅呼吁大众停止将AI生成内容视为粗制滥造之物，更希望科技行业不再将AI描绘成人类的替代者。他期盼整个行业能开始将其定位为辅助人类的生产力工具。

然而这种框架存在根本矛盾：当前多数AI产品的营销策略恰恰以替代人力作为定价依据和成本合理性证明。与此同时，AI领域的领军人物不断发出警告，称这项技术将很快导致大规模失业。例如今年五月，Anthropic首席执行官达里奥·阿莫代伊预警AI可能取代半数初级白领岗位，未来五年失业率或将攀升至10%-20%，上月他在《60分钟》节目中再次强调了这一观点。

加入Disrupt 2026候补名单
当早鸟票开售时，加入候补名单可优先获取门票。往届Disrupt大会曾汇聚谷歌云、Netflix、微软、Box、Phia、a16z、ElevenLabs、Wayve、Hugging Face、埃拉德·吉尔和维诺德·科斯拉等250多位行业领袖，举办200多场助力企业成长与提升竞争力的会议。更有数百家跨领域创新初创企业等待您的对接。

不过这些末日预言的真实性尚待验证。正如纳德拉暗示的，当前大多数AI工具并未取代劳动者，而是被劳动者所使用（前提是人类不介意核查AI工作的准确性）。麻省理工学院正在进行的"冰山项目"研究显示，AI目前能够承担约11.7%的人类有偿劳动。虽然媒体普遍将其解读为"AI可替代近12%工作岗位"，但该项目强调实际测算的是"工作中可交由AI处理的比例"，并据此计算对应薪资。值得注意的是，研究列举的案例包括护士的自动化文书工作与AI编写的程序代码。

这并非否认AI对某些职业的冲击。据《机器中的鲜血》专栏指出，企业平面设计师和营销博主是受影响较大的群体，应届初级程序员的失业率也居高不下。但另一方面，掌握AI工具的高技能艺术家、作家和程序员确实能产出更优质的作品。目前AI尚无法取代人类的创造力。

随着2026年临近，新数据显示AI渗透最深的领域反而呈现繁荣态势。先锋集团2026年经济预测报告发现："约100个最易受AI自动化影响的职业，在就业增长和实际工资涨幅方面反而领先劳动力市场其他领域。"报告结论指出：精通AI运用者正在提升自身价值而非被替代。

颇具讽刺意味的是，微软自身去年的行为助长了"AI取代人力"的叙事。尽管截至6月的财年创下营收利润纪录，该公司仍在2025年裁员超1.5万人，并将AI成功应用列为裁员原因之一。纳德拉甚至在财报发布后就此发表公开备忘录。值得玩味的是，他并未直言内部AI效率提升导致裁员，但强调微软必须"重新构想新时代的使命"，并将"AI转型"列为该时期三大业务目标之一（另两项为安全与质量）。

2025年AI相关失业的真相更为复杂。正如先锋报告所指，这更多与常规商业实践相关——例如从增长放缓领域撤资转向新兴领域——而非内部AI效率提升。客观而言，追逐AI浪潮中裁员的企业不止微软。据CNBC报道，挑战者·格雷·圣诞公司的研究显示，2025年美国近5.5万裁员被归因于AI技术，去年亚马逊、Salesforce等追逐AI的科技公司都进行了大规模裁员。

平心而论，对于那些花费过多时间在社交媒体上浏览表情包和AI生成短视频的用户而言，"粗制滥造的内容"或许正是AI最有趣（即便不是最佳）的用途之一。

英文来源：

A couple of weeks after Merriam-Webster named “slop” as its word of the year, Microsoft CEO Satya Nadella weighed in on what to expect from AI in 2026.
In his classic, intellectual style, Nadella wrote on his personal blog that he wants us to stop thinking of AI as “slop” and start thinking of it as “bicycles for the mind.”
He wrote, “A new concept that evolves ‘bicycles for the mind’ such that we always think of AI as a scaffolding for human potential vs a substitute.”
He continued: “We need to get beyond the arguments of slop vs sophistication and develop a new equilibrium in terms of our ‘theory of the mind’ that accounts for humans being equipped with these new cognitive amplifier tools as we relate to each other.”
If you parse through those syllables, you may see that he’s not only urging everyone to stop thinking of AI-generated content as slop, but also wants the tech industry to stop talking about AI as a replacement for humans. He hopes the industry will start talking about it as a human-helper productivity tool instead.
Here’s the problem with that framing, though: Much of AI agent marketing uses the idea of replacing human labor as a way to price it, and justify its expense.
Meanwhile, some of the biggest names in AI have been sounding the alarm that the tech will soon cause very high levels of human unemployment. For instance, in May Anthropic CEO Dario Amodei warned that AI could take away half of all entry-level white-collar jobs, raising unemployment to 10-20% over the next five years, and he doubled down on that last month in an interview on 60 Minutes.
Join the Disrupt 2026 Waitlist
Add yourself to the Disrupt 2026 waitlist to be first in line when Early Bird tickets drop. Past Disrupts have brought Google Cloud, Netflix, Microsoft, Box, Phia, a16z, ElevenLabs, Wayve, Hugging Face, Elad Gil, and Vinod Khosla to the stages — part of 250+ industry leaders driving 200+ sessions built to fuel your growth and sharpen your edge. Plus, meet the hundreds of startups innovating across every sector.
Join the Disrupt 2026 Waitlist
Add yourself to the Disrupt 2026 waitlist to be first in line when Early Bird tickets drop. Past Disrupts have brought Google Cloud, Netflix, Microsoft, Box, Phia, a16z, ElevenLabs, Wayve, Hugging Face, Elad Gil, and Vinod Khosla to the stages — part of 250+ industry leaders driving 200+ sessions built to fuel your growth and sharpen your edge. Plus, meet the hundreds of startups innovating across every sector.
Yet we currently don’t know how true such doomsday stats are. As Nadella implies, most AI tools today don’t replace workers, they are used by them (as long as the human doesn’t mind checking the AI’s work for accuracy).
One oft-cited research study is MIT’s ongoing Project Iceberg, which seeks to measure the economic impact on jobs as AI enters the workforce. Project Iceberg estimates that AI is currently capable of performing about 11.7% of human paid labor.
While this has been widely reported as AI being capable of replacing nearly 12% of jobs, the Project says what it’s actually estimating is how much of a job can be offloaded to AI. It then calculates wages attached to that offloaded work. Interestingly, the tasks it cites as examples include automated paperwork for nurses and AI-written computer code.
That’s not to say there are no jobs being heavily impacted by AI. Corporate graphic artists and marketing bloggers are two examples, according to a Substack called Blood in the Machine. Then there are the high unemployment rates among new-grad junior coders.
But it’s also true that highly skilled artists, writers, and programmers produce better work with AI tools than those without the skills. AI can’t replace human creativity, yet.
So it’s perhaps no surprise that as we slide into 2026, some data is emerging that shows the jobs where AI has made the most progress are actually flourishing. Vanguard’s 2026 economic forecast report found that “the approximately 100 occupations most exposed to AI automation are actually outperforming the rest of the labor market in terms of job growth and real wage increases.”
The Vanguard report concludes that those who are masterfully using AI are making themselves more valuable, not replaceable.
The irony is that Microsoft’s own actions last year helped give rise to the AI-is-coming-for-our-jobs narrative. The company laid off over 15,000 people in 2025, even as it recorded record revenues and profits for its last fiscal year, which closed in June — citing success with AI as a reason. Nadella even wrote a public memo about the layoffs after these results.
Notably, he didn’t say that internal AI efficiency led to cuts. But he did say that Microsoft had to “reimagine our mission for a new era” and named “AI transformation” as one of the company’s three business objectives in this era (the other two being security and quality).
The truth about job loss attributed to AI during 2025 is more nuanced. As the Vanguard report points out, this had less to do with internal AI efficiency and more to do with ordinary business practices that are less exciting to investors, like ending investment in slowing areas to pile in to growing ones.
To be fair, Microsoft wasn’t alone in laying off workers while pursuing AI. The technology was said to be responsible for almost 55,000 layoffs in the U.S. in 2025, according to research from firm Challenger, Gray & Christmas, CNBC reported. That report cited the large cuts last year at Amazon, Salesforce, Microsoft, and other tech companies chasing AI.
And to be fair to slop, those of us who spend more time than we should on social media laughing at memes and AI-generated short-form videos might argue that slop is one of AI’s most entertaining (if not best) uses, too.