Generative AI

Overview

Hideki Saito adopts a supportive stance on generative AI, actively using and publishing works created with tools like Avalab. While he strives to understand perspectives from both proponents and critics, he expresses frustration with sentiment-based arguments from opponents, often finding them lacking in substantive or constructive criticism.

Articles

• AI Agents: Just a FUD? - The article critiques the practicality of AI agents, arguing that while they may be useful in niche applications, their widespread adoption faces significant hurdles. Unlike traditional automation, AI agents are designed to operate independently, but their inherent unpredictability makes them unreliable for high-stakes fields like healthcare and finance. The probabilistic nature of AI, susceptibility to data manipulation, and challenges with real-time voice interactions further complicate their effectiveness. Additionally, the lack of clear accountability for AI-driven decisions raises ethical and legal concerns. While AI agents may succeed in controlled environments where errors have minimal impact—such as logistics or simulations—the vision of fully autonomous AI handling critical tasks remains unrealistic. The article concludes that AI agents are more of a speculative idea than a viable replacement for human decision-making in most fields.
• Beyond the Buzzwords: Rethinking AI in Writing - The article challenges the oversimplified assumptions about AI-generated writing, particularly the idea that words like “Furthermore” or polished text automatically indicate AI involvement. It highlights how writing style is influenced by reading habits, education, and editing tools like Grammarly, making AI detection based on superficial markers unreliable. The piece also critiques the flawed logic of AI “policing” in creative fields and argues that AI can be a helpful tool rather than a threat. Ethical concerns arise when anti-AI articles rely on sensationalism rather than meaningful discussion. Instead of fixating on AI detection, the article advocates for evaluating writing based on purpose, accuracy, and value, encouraging a more constructive dialogue on human-AI collaboration.
• The State of Generative AI Sentiment in Japan’s Illustration Scene – This article examines the polarized debate surrounding generative AI in Japan’s illustration scene. It analyzes the cultural and ethical concerns fueling the divide between traditional artists and those embracing AI tools, highlighting the impact on business practices and the challenges of finding common ground in a rapidly changing creative landscape.
• The Tricky Terrain of Defining “AI Involvement” in Writing – This article explores the blurred lines between human and AI contributions in writing. It examines the spectrum of AI assistance, from grammar correction to full text generation, and the difficulties in detecting such involvement. The ethical considerations of disclosure and the impact on perceived authenticity are discussed. The article concludes by advocating for transparency and prioritizing the inherent value of the writing, irrespective of the tools employed.
• The Alarming Assumption: When Writing Gets Dismissed as AI-Generated – This article discusses the author’s personal experiences as a non-native English speaker whose writing style is sometimes mistaken for AI-generated. It highlights the challenges faced by non-native writers and criticizes the assumption that polished or unusual writing is automatically AI-produced. The author advocates for appreciating the diversity of voices and judging writing by its content rather than its perceived origin.