Recently, I came across a post from Sarah Friar, the CFO of OpenAI, mentioning the possibility of a $2000/month subscription for their AI product—a tool so advanced that it might serve as a “replacement” for human employees due to its PhD-level intelligence. This sparked a wave of reactions, with many asserting that AI is destined to replace human workers entirely.
(Ref: https://x.com/tsarnick/status/1868201597727342941?s=46)
From my perspective, based on my personal experience, the reality is a little bit different.
AI is undoubtedly a powerful tool. Its ability to process vast amounts of data, generate solutions, and even create content is remarkable. However, despite its intelligence, AI is not infallible. Mistakes still happen—sometimes subtly, other times glaringly—and these errors often require human intervention to correct or refine the outcomes. Without this critical human oversight, the results produced by AI might fail to achieve the intended goals.
Take software development as an example. AI can efficiently generate code for a standard website with basic functionalities. It can assist in tasks like debugging or optimizing algorithms. However, when it comes to building a brand-new platform with complex features tailored to solve intricate problems—such as those found in semiconductor manufacturing or banking anti-money laundering systems—the situation changes. These domains demand a deep understanding of industry-specific challenges, regulatory frameworks, and innovative problem-solving approaches that, at least for now, only human professionals can provide.
If we’re talking about AI as a “replacement,” it mostly applies to basic human tasks. For instance, a lawyer’s assistant might collect and organize materials before tackling advanced legal work, and AI could potentially perform this task better than a human assistant. However, AI cannot yet craft an ideal legal brief with the nuance and strategic thinking of a skilled lawyer.
Therefore, rather than viewing AI as a replacement, I prefer to see it as an “assistant.” AI is not the enemy of the labor force but a powerful tool that enhances our efficiency and effectiveness. It can support our work, allowing us to focus on tasks that truly require human expertise.
Of course, we know AI is advancing rapidly and may replace more roles in the near future. But for now, I see AI as a helpful assistant—a tool that I rely on while ensuring the final decisions and outcomes remain under my control to guarantee success.
Maybe I’m wrong, but what do you think? 🙂
最近，我看到一則來自 OpenAI 的 CFO Sarah Friar 的文章，提到他們的 AI 產品可能推出每月 2000 美元的訂閱服務——這是一款如此先進的工具，被認為可以憑藉其相當於博士水準的智慧作為「人類員工的替代品」。這引發了廣泛的討論，許多人認為 AI 註定會完全取代人類勞動力。
（Ref：https://x.com/tsarnick/status/1868201597727342941?s=46）
從我的觀點來看，根據我的個人經驗，現實情況有些不同。
AI 無疑是一個強大的工具。它處理大量數據、生成解決方案，甚至創建內容的能力非常出色。然而，儘管 AI 非常智能，但它並非萬無一失。錯誤仍然會發生——有時是細微的，有時是明顯的——這些錯誤往往需要人類的介入來修正或完善結果。如果缺乏這種關鍵的人類監督，AI 產生的結果可能無法達成預期目標。
以軟體開發為例。AI 可以高效且快速地生成具有基本功能的標準網站程式碼，也可以協助完成調試或優化演算法等任務。然而，當涉及到構建一個全新且具備複雜功能的平台，目標是要解決諸如半導體製造或銀行反洗錢系統這樣的複雜問題時，情況就會截然不同。這些領域需要對特定行業知識、監管框架以及創新解決方案有深入的理解，而目前，這些能力只能由專業的人類工作者所提供。
如果我們談論 AI 作為「替代品」，它大多數只能替代基本的人類工作。例如，一位律師的助理可能會在處理高級法律工作之前收集並整理材料，而 AI 有可能比人類助理更有效地完成這項任務。然而，AI 尚無法撰寫一份具有答辯策略和縝密思維的理想法律文件，而這卻是人類律師的強項。
因此，與其將 AI 視為替代品，我更願意將其視為「助理」。AI 不是人類勞動力的敵人，而是一個增強我們效率和效能的強大工具。它可以支援我們的工作，讓我們能專注於真正需要人類專業知識的任務。
當然，我們知道 AI 正在快速進步，未來可能會取代更多的工作崗位。但就目前而言，我將 AI 視為一位得力助手——一個我依賴的工具，同時確保最終決策和結果由我掌控，從而保證最終的成功。
但這是我個人的想法，也許我錯了，想請教你怎麼看呢？ 🙂