Over the past few months, my thinking has elevated through several levels. There were so many ideas I wanted to record, but due to the time pressure of my thesis and various chores, they were lost—a pity. Thinking slices are like that; if you’re a day late, everything has changed. This time, after watching Jaron Lanier’s speech and finishing a meeting, I quickly sliced off this biased piece of thought and memory. At the moment of writing, I haven’t yet submerged into that deep sea of thought; therefore, compared to the intricate cognitive links in the first piece “Preface?”, this one is relatively plain and direct.
Skills#
For the second piece, I’ll take “Skill” (or Skills?), a recurring theme lately, as the subject. But instead of discussing “Skills” in the AI circle, let’s talk about human skills.
Skill has successfully lowered the technical threshold of Context Design to a level understandable by ordinary people, providing a universal standard format that is more intuitive than MCP. At the same time, OpenClaw (Clawdbot) has also introduced AI Agents into people’s lives in the most “approachable” way. The synergy between these two, which are closest to people’s existing skills, has led to a constant stream of repetitive rhetoric since the 2026 New Year, which is somewhat nauseating. There’s not much depth to discuss; it’s still struggling within the framework predictable since 2022. Reflecting on why I feel indifferent to this, my preliminary conclusion: “I can’t learn any new skills.”
“Human-Centric” is a keyword that has most deeply influenced my thinking and research recently—centering on humans, based on humans. In the fields of HCI, architecture, and various design disciplines, this is an extremely fundamental perspective: discussing how a space is needed by people, how people interact with UI, etc. The focus is always on how the designed product must meet certain people’s needs. However, AI is different from previous technologies. Engines, computers, the internet, and smartphones “passively” influence human behavior, while AI “relatively actively” influences users. The truth is that AI has already moved beyond the scope of passively catering to needs; AI can autonomously adjust and act to satisfy user needs. This is the foreseeable future. In other words, the discussion should no longer be limited to the surface of needs, but should focus on what “bonuses” AI brings to users beyond the product itself.
Empowerment? Disempowerment?
AI-Empower, AI empowerment—I’m not sure if this is an AI feature word, but in any case, most people who use this term base their discussion on AI lowering the professional technical threshold, allowing anyone to easily obtain products that previously belonged to specialized professional fields, like Vibe Coding Revit tools or automated real estate status analysis. This further extends into discussions about the transformation of professionalism: “In the future, humans should value the ability to choose and make decisions, becoming Curators.” The key is how to select good solutions and good results. However, another group of people proposes the opposite view. When tools can directly generate results and humans are left with only prompts and choices, will other abilities gradually erode?
“People degrade themselves in order to make machines seem smart all the time.” — Jaron Lanier, You Are Not a Gadget
Facing this issue, I believe the discussion should be elevated to another level. The key is not the skill itself, but the new perspective that comes with the skill. Does this technology “give me a new pair of eyes”? Looking back at the previous discussion from this perspective, Vibe coding will not be limited to improving efficiency and earning short-term profits, nor just outputting “products.” It’s about understanding the logic and decisions behind the generation process, forcing oneself to switch thinking perspectives; or the product itself is used to extend one’s vision, such as literature reviews, data analysis, etc. When users consciously use AI to augment their vision, I think the current discussion about AI will naturally jump out of the “Skill” framework.
If we discuss the most lacking skill currently, I believe it is “true” interdisciplinary thinking. There are too many scholars today who think they are interdisciplinary, but what they talk about is merely the most superficial intersection of two existing fields. They use rigid, traditional thinking to discuss, comment on, and even teach interdisciplinarity, without realizing they are just standing in their own small box borrowing terminology from other fields—very regrettable. Perhaps in the future, there should exist some kind of academic hierarchy for interdisciplinary thinking? Integrating the thinking logic of multiple fields to develop personal value is, I think, one of the most critical skills in the current AI era. Let’s encourage each other.
“We have modified our environment so radically that we must now modify ourselves in order to exist in this new environment. We can no longer live in the old one.” — Norbert Wiener, The Human Use of Human Beings: Cybernetic and Society, 1954
Finally, I’ll share a skill I’ve been learning lately: “Emptying the mind” (Spacing out).
Blessings.
“David might make mistakes, please verify.”
