# Quality Articoli collegati

Il Centro Notizie HTX fornisce gli articoli più recenti e le analisi più approfondite su "Quality", coprendo tendenze di mercato, aggiornamenti sui progetti, sviluppi tecnologici e politiche normative nel settore crypto.

Those Who Use AI to 'Quickly Finish the Job' Are Ruining the Workplace

Some people are using AI tools to quickly produce superficially complete work, like reports, proposals, or performance reviews, just to "check a box." This results in "Workslop" – workplace AI output that looks polished but lacks factual verification, professional judgment, and real value. The core problem is the shifting of work costs: while the generator saves time, the receiver must spend hours verifying information, correcting errors, and deciphering vague conclusions. This undermines team efficiency and trust. Examples include a manager using AI for vague, unactionable employee feedback, an analyst generating weekly reports with misleading AI-interpreted data, and a team producing extensive project documents with no one able to justify the underlying assumptions. The issue isn't AI itself, but how it's used. It amplifies existing work habits. Responsible users treat AI as an assistant for drafting and organizing, then add firsthand verification and judgment. Those merely looking to "hand in" work delegate responsibility to the AI and their colleagues. Ultimately, the value of work should be judged not by its speed or polish, but by whether it provides new facts, clear judgments, and an accountable owner. AI cannot understand context, make critical decisions, or bear the consequences – those remain irreplaceably human responsibilities.

marsbit07/29 08:11

Those Who Use AI to 'Quickly Finish the Job' Are Ruining the Workplace

marsbit07/29 08:11

Li Shanyou: The World Is Getting More Competitive, But We Can Choose to Be Slower, Go Deeper, and Turn Inward

**Title:** Li Shanyou: The World Is Moving Faster and More Chaotically; We Can Choose to Be Slower, Deeper, and More Inward **Summary:** In an era where AI evolves weekly and past experiences often fail, the article argues for returning to First Principles. It shifts focus from "cognition" (knowledge/content), which AI has disrupted, to "consciousness" or "the capacity to know"—the foundational ability that enables cognition and determines its scope and depth. The piece uses Zhang Yiming (founder of ByteDance) as a case study for the "second-order consciousness" of the mobile internet era: success through rational logic, modeling, and deduction, which allowed him to surpass earlier "first-order" entrepreneurs reliant on copying experiences. However, the author suggests that the AI era demands a further leap to "third-order consciousness"—operating from a place of inner inspiration, vision, or "ideas," beyond pure rationality. Figures like Elon Musk and Steve Jobs are cited as examples, with DeepSeek's Liang Wenfeng presented as a potential representative of this new, AI-native wave of entrepreneurs driven by curiosity and innovation rather than just commercial logic. For the majority who may not become such visionary leaders, the author proposes an alternative path: the "Path of Quality" or "Work as Art." This involves turning inward, focusing on the immediate task at hand, and imbuing one's work with care and dedication—creating a "quality" product not as a means to an end, but as an end in itself. This process refines both the work and the individual's own consciousness. In a hyper-competitive, fast-paced world, this path offers a counterpoint: to be slower, go deeper, and focus inward, finding meaning by creating quality in everyday endeavors, regardless of scale.

marsbit07/17 04:13

Li Shanyou: The World Is Getting More Competitive, But We Can Choose to Be Slower, Go Deeper, and Turn Inward

marsbit07/17 04:13

Slow Down: The Answer in the Age of Agents

In the era of generative AI, the software industry is shifting from amazement to efficiency anxiety. However, as coding agents are increasingly used in production, issues like amplified errors, uncontrolled complexity, and reduced system reliability emerge. The author argues that agents lack human-like learning from mistakes and, without proper bottlenecks and feedback, minor issues quickly escalate. Their limited perspective and low recall in complex codebases further worsen structural chaos. The core problem isn’t the technology but the premature surrender of human judgment and control driven by anxiety. Instead of fully outsourcing work to agents, the author advocates for a balanced approach: assign agents localized, well-defined tasks while retaining human oversight for system design, quality assurance, and critical decisions. Slowing down becomes a strength—it ensures understanding, enables informed trade-offs, and maintains control. Ultimately, what’s scarce in the AI age isn’t faster code generation but the judgment to manage complexity and the discipline to choose quality over speed.

marsbit03/29 15:35

Slow Down: The Answer in the Age of Agents

marsbit03/29 15:35

AI Agent Outputs Garbage? The Problem Is You're Not Willing to Burn Enough Tokens

The core argument is that the quality of an AI Agent's output is directly proportional to the number of tokens invested in the process. More tokens lead to fewer errors, as they allow for deeper reasoning, multiple independent attempts, self-critique from fresh contexts, and verification through testing. This approach can solve problems of scale and complexity but fails when facing novel problems not present in the model's training data. For such novel challenges, human domain knowledge and guidance are essential. Two practical, immediate solutions are proposed: implementing an automatic review cycle (WAIT) for the Agent to repeatedly critique and fix its work, and establishing frequent verification checkpoints (VERIFY) where a separate Agent validates outputs to catch errors early. The key takeaway is that insufficient token investment is often the primary reason for poor Agent performance, not the underlying framework.

marsbit03/23 06:14

AI Agent Outputs Garbage? The Problem Is You're Not Willing to Burn Enough Tokens

marsbit03/23 06:14

In a World of Dramatic Change, How Should Humanities Workers Better Use AI?

In a rapidly changing landscape, humanities professionals are increasingly turning to AI not as a magic solution, but as a practical tool integrated into their research, writing workflows. This guide outlines key principles for effectively using AI, moving beyond simple "prompts" to a systematic, controllable methodology. The approach is built on three core tenets: processes must be traceable, verifiable, and supervised; the user must remain in control; and the final output must be something the creator is willing to sign their name to. Key principles include: * **Treat AI as a workbench, not a wish-granter:** Clearly define tasks, audiences, and standards instead of making vague requests. * **You are the responsible agent:** Provide clear context, constraints, and executable steps. Dissatisfaction often stems from unclear instructions, not AI failure. * **Compare multiple models:** Different AIs have different strengths (writing, reasoning, coding); use them like a team. * **Manage expectations:** Assume AI has the knowledge level of a top undergraduate; provide examples and standards for specialized tasks. * **Break tasks into steps:** A white-box process of small, reliable steps is better than a single, error-prone black-box request. * **Industrialize first, then automate:** Define and structure your workflow into reproducible steps before assigning sub-tasks to AI. * **Anticipate AI's laziness:** Remove format barriers (e.g., clean text from PDFs/websites) to focus its effort on comprehension. * **Prioritize compression over expansion:** It's more reliable to condense large amounts of provided material than to ask AI to generate content from little context. * **Iterate on the pipeline, not the output:** Aim for a system that consistently produces good-enough drafts (e.g., 75/100) rather than manually perfecting each result. * **Generate quantity to find quality:** Request multiple versions (e.g., 5 summaries, 50 headlines) to combat mediocrity and discover excellent samples. * **Act as a head chef:** Provide clear feedback for revisions instead of rewriting the output yourself. The ultimate quality of work depends on **materials × taste**. AI enhances interaction with materials, but genuine research, unique sources, and cultivated judgment remain irreplaceable. The goal is to replace anxiety with practical skill by engineering tasks, making processes transparent, and integrating AI as a verb within a credible,署名-worthy creative process.

marsbit03/05 05:20