现在数据讲述了一个不同的故事。The tool that workers once raced to adopt covertly has become, for a large and growing share of the workforce, the tool they’ve stopped using altogether.不是因为它不起作用。Because they’re afraid of what happens when it works too well.

一项针对 14 个国家/地区 3,750 名高管和员工的新全球调查由SAPsubsidiary WalkMe for its fifth annual State of Digital Adoption report, finds that more 54% of workers bypassed their company’s AI tools in the past 30 days and completed the work manually instead.另外 33% 的人根本没有使用过人工智能。加起来，大约十分之八的企业员工正在回避或积极拒绝雇主花费创纪录的资金来部署的技术。Average digital transformation budgets rose 38% year-over-year to $54.2 million — yet 40% of that spend has been underperforming due to adoption failures.

高管们对员工的真实感受视而不见

早期的热情所掩盖的东西现在在数字中可见一斑。只有 9% 的员工相信人工智能可以做出复杂的关键业务决策，而高管的这一比例为 61% — 存在 52 点信任鸿沟。88% 的高管表示，他们的员工拥有足够的工具；only 21% of workers agree — a 67-point gap on tool adequacy alone.Executives and their employees are, in the report’s language, “describing fundamentally different companies.”

怀疑论者也有数据支持。史蒂夫·汉克约翰·霍普金斯大学的经济学家已经经历了足够多的技术周期，从内部了解炒作是什么样的。“人工智能没有兑现，”他告诉我们财富最近。欢迎来到现实世界。Forget the AI bubble.你知道，它没有兑现。You look at all the surveys and yeah, everybody’s using it a little bit, but you dig into it and it hasn’t done much.” Hanke’s bottom line: “Productivity, by the way, it was weak.如果人工智能得以实现，生产力将会大幅提高。You listen to these Silicon Valley guys and they say we’re gonna have GDP going to 5% of 6%.生产力将提高到六。这只是没有发生。”

这种怀疑论本身与 WalkMe 数据的发现是一致的。丹·阿迪卡WalkMe 的首席执行官兼联合创始人一直在一线追踪这种分歧。他定期与首席信息官会面，并向他们提出一个简单的问题：有多少人真正使用人工智能来做有意义的工作？“这个数字不到 10%，”他说。

阿迪卡用了一个比喻，受到这位特定编辑的青睐同样，人工智能的速度就像一辆跑车。他说，他最喜欢的比喻是，如果你给每个员工买一辆跑车，但他们不知道如何驾驶它——他们不具备人工智能技能。

部分问题是结构性的，而不是行为性的。“You buy every employee that sports car, the Ferrari, but they don’t know how to drive,” Adika said.“They don’t have fuel sometimes, which is the context.知道如何驾驶就是提示。And in some cases, there are not even enough roads — there’s no API or MCP server to actually do what you want to do.” What do you do when you have a Ferrari, but no driver, no fuel, and no roads?你走得不是很快。

布拉德·布朗毕马威美国税务技术与创新全球主管在另一次采访中使用了几乎完全相同的比喻财富。“It’s like an F1 car driver,” he said.– F1 赛车太棒了。但如果你没有技术精湛、才华横溢的司机，那么这个工具对你的帮助不大。事实上，两位经验丰富的技术专家（一位是创始人，一位是四大合伙人）自发地得出了相同的描述，这表明他们正在描述他们都曾多次亲眼目睹的东西。

鸿沟正在让企业付出代价

现在，无动力法拉利的下游成本是可以量化的。WorkMe 报告发现，员工每年因技术摩擦而损失相当于 51 个工作日（近两个月），比 2025 年增加了 42%。即每周 7.9 小时。高盛经济学家本周报告称，正确使用人工智能的员工平均每天可以节省 40 到 60 分钟的时间。数学几乎是对称的：人工智能为那些善于使用它的人带来的生产力几乎完全等于它对那些无法让它发挥作用的人所破坏的生产力。

古老的影子人工智能故事在表面之下仍然存在。78% 的高管表示，他们希望规范影子人工智能的使用，但只有 21% 的员工表示曾收到过有关人工智能政策的警告，34% 的员工甚至不知道雇主批准了哪些工具。Executives are threatening punishment for behavior that they’ve never explained is prohibited.这种矛盾如此深刻，以至于 62% 的高管私下承认，与根本不利用人工智能的风险相比，未经批准的影子人工智能的风险被夸大了。

“使用影子人工智能并不是一种惩罚行为，而是一个解决系统性差距的机会，”Futurum Group 副总裁兼企业软件数字工作流程研究总监 Keith Kirkpatrick 表示。– 当员工使用未经批准的人工智能工具时，他们是在弥补受认可的工具和不明确的治理所留下的绩效或效率差距。 –

人工智能脱离

What’s new — and what the data is only beginning to capture — is the layer beneath shadow AI.Workers who aren’t sneaking around the rules.没有做任何事的工人。

Adika was asked what he’d call this dynamic.他停了下来。“They have pride in what they do,” he said, about workers who are resisting AI adoption.“他们不会让某些人工智能机器人接管，他们总是会发现并展示该工具与他们相比的缺陷。”听起来毫无疑问，就像安静地戒烟— the pandemic-era phenomenon in which workers stopped going above and beyond without formally resigning.这也可能是对人工智能工具的一种非常可以理解的挫败感，它们只是不停地产生幻觉，浪费了他们承诺节省的时间。

“The organizations that get this right won’t be the ones that just automated the most tasks,” Adika said.“They’ll be the ones that figured out when the human should act, when the agent should act, and how the handoff between them works.这种交接就是信任的所在。And right now, most companies haven’t even started thinking about it.” To this point, t麻省理工学院的研究发现，90% 的员工仍然更喜欢由人类来从事关键任务工作，显然不愿意深入研究。

甲骨文继发布类似声明后，该公司宣布裁员数万名员工块, although critics see this as “AI washing,” or disguising over-hiring with a convenient excuse that happens to boost the stock price.普通民众并没有失去逻辑。“We will be in a certain point of time when we will feel uncertainty, fear, we’ll see layoffs,” Adika said.“So I think it’s kind of a transition period that will happen over time.

But again, at the end of the day, people are not using it yet.” Adika was also clear that workers staying away from AI are not wrong to sense something real — they’re wrong about the conclusion.“你不会看到任何一家银行或保险公司的首席执行官明天就离职并解雇很多人，因为谁来做这项工作？”他说，他看到了一个“大问题”即将到来，因为声称人工智能将取代所有人的说法将不得不面对这样一个事实：“这只是现在还没有发生。”

熟练驾驶员问题

Brown said he’s spending more time than ever thinking about what it actually takes to close the gap between the Ferrari and the driver.在毕马威会计师事务所，他开始将劳动力分为构建者、创客和高级用户——人工智能能力的不同层次，并附加了明确的职业道路。“Our focus right now is to craft incentives and career paths to get all our people to that level,” he said.“现在是人类赶上技术进步的时候了。”

The critical insight in that framing is that the problem isn’t intelligence, nor is it even training in the traditional sense.“我认为，凭借批判性思维和判断力方面的人类技能，”布朗说，“这将帮助人们成为创客”，即能够灵活利用人工智能工具的工人，包括使用它们自己构建新工具。在他看来，面临最大风险的工人并不是那些缺乏技术技能的工人。They’re the ones whose employers haven’t given them a safe space, a path, or an incentive to try.

三分之一的企业员工从未使用过人工智能工具，他们表示支持水平最低，培训最少，对中断的焦虑程度最高。WalkMe 报告仔细指出，他们并不抵制人工智能。他们根本就没有达到。As to whether the evolution of these tools is outpacing workers’ ability to catch up, Brown acknowledged that he definitely feels a gap.

进化是可能的——而且很重要

一旦汉克弄清楚自己想用人工智能做什么，他就能重新振作起来。“人工智能对我来说有点像另一个研究助理，”他说，“它节省了大量时间，因为如果我有一个研究助理做这些事情，我就必须把他们送到图书馆。”They’d be screwing around over there for a week doing something I can do on AI in about an hour.” The caveat: “You have to know what they’re good for.” And, crucially, you have to know enough about the subject matter to catch the errors. “I know what to ask AI.I know how to structure what I want done,” Hanke said, pointing to his decades of domain expertise across economics, commodities, and international finance.

His own trajectory — from outright banning student use to cautious skepticism to daily reliance — tracks the arc many serious thinkers have traveled. He said he went from “‘no’, to ‘maybe’, to ‘this is great—but some of these tools suck.'” His verdict on the tools themselves is characteristically blunt: “There are all kinds of AI.其中一些确实很糟糕。这取决于您的需要。 –

Brown’s view is that this is ultimately an optimistic story — but only for those who move.“The winners are the ones where you have your workforce effectively leveraging the capabilities of AI,” he said.– 不倾向于人工智能的劳动力将会受到挑战。And a work environment that is overly oriented to AI without the value of the human workforce is going to struggle.”