本站提供正體中文版。切換到正體中文本站提供简体中文版。切换到简体中文このサイトには日本語版があります。日本語で表示이 사이트는 한국어로도 제공됩니다.한국어로 보기Diese Website ist auch auf Deutsch verfügbar.Auf Deutsch ansehenEste sitio web también está disponible en español.Ver en españolQuesto sito è disponibile anche in italiano.Visualizza in italianoCe site est également disponible en français.Afficher en françaisEste site também está disponível em português.Ver em portuguêsDeze website is ook beschikbaar in het Nederlands.In het Nederlands bekijkenЭтот сайт также доступен на русском языке.Смотреть на русскомयह वेबसाइट हिन्दी में भी उपलब्ध है।हिन्दी में देखेंهذا الموقع متاح أيضًا باللغة العربية.عرض بالعربيةSitus ini juga tersedia dalam bahasa Indonesia.Lihat dalam bahasa IndonesiaBu site Türkçe olarak da mevcut.Türkçe görüntüleTa strona jest dostępna także po polsku.Wyświetl po polskuTrang web này cũng có phiên bản tiếng Việt.Xem bằng tiếng Việtاین وب‌سایت به فارسی هم در دسترس است.مشاهده به فارسی

Facing the AI Era, My Feelings Are Always Mixed

AI2026.02

Facing the AI era, my feelings are always mixed.

The fascinating side is that there’s forever an endless supply of new knowledge before me, so much that one person’s time simply can’t hold such a vast volume of information; but the exasperating side is that AI also often leads us on detours—it’ll pretend to understand what it doesn’t, produce hallucinations, and even fail to grasp certain specific operations and meanings with any precision.

Gold and gray sand trickling slowly down through an hourglass, against a backdrop of intricate clockwork gears

A Whole Day, Four-Fifths of It Spent on Trial and Error

Today I spent a great deal of time feeling my way through a new tool, in order to retire an old method I’d grown accustomed to over the past twenty years but that no longer meets today’s needs. It was a transition I had no choice but to face.

Looking back over the whole day, roughly four-fifths of the time went into trial and error and groping around, and only the crucial one-fifth was the part that actually produced results. Although I hit my target smoothly in the end and did save considerable time for the future, I couldn’t shake the feeling that those four-fifths of effort had been spent somewhat in vain.

If Only We Could Save the Cost of the Detours

I can’t help thinking: if we could save all these detour costs and pour them entirely into productivity, then the knowledge we could absorb would surely grow several times over.

Perhaps this is simply the limitation of AI at this stage. It’s already powerful enough that you can’t do without it, yet not yet mature enough for you to trust it completely. But looked at from another angle, this “not quite there yet” state also makes one all the more eager for its future evolution and maturity.

Contradictions aside, the road still has to be walked forward.