Every day we ask AI to draft, summarize, and explain. But when does a chatbot’s answer become a document you can trust? Drawing on a conversation with ChatGPT, Sascha Donner shows how humans confer the trust, force, and answerability that turn an AI answer into an accountable document.
Read More
Based on a conversation in the InfoFire series between computer scientist Ricardo Baeza-Yates and information scientist Shalini R. Urs, this article explores the philosophical, cognitive, and societal shifts that have accompanied generative AI’s evolution from an external tool to an internal “cognitive companion.”
It essentially outlines five core arguments regarding how AI is reshaping human thought:
• The Great Illusion (Imitation vs. Comprehension): This section highlights the “ELIZA Effect,” in which people mistake generative AI’s fluent, polished language for genuine understanding. Baeza-Yates prefers the term “Computational Intelligence” to “Artificial Intelligence,” arguing that large language models generate language statistically. Seduced by linguistic fluency, we increasingly confuse prediction with cognition.
• The Delegation of Mind (Cognitive Offloading): Unlike older technologies that automated physical tasks or basic mathematics, generative AI risks “cognitive substitution” or “cognitive surrender.” By outsourcing intellectual exercises such as writing, coding, and reasoning to AI, humans may lose the productive cognitive struggle required to form original thoughts and develop critical judgment.
• Truth in the Age of Prediction: Information retrieval, such as through traditional search engines, preserves human agency and verification, whereas AI generation encourages immediate acceptance. LLMs do not “hallucinate” out of malfunction; they “confabulate” because their objective is statistical probability rather than factual truth. Detaching credibility from reality creates an “era of algorithmic plausibility.”
• Intelligence without Experience: AI systems process language but lack physical embodiment, causal reasoning, common sense, and “tacit knowledge” gained from living in the physical world. Furthermore, because training data relies heavily on dominant digital structures, such as English, it creates “AI colonialism”—a quiet standardization that compresses global cultural and intellectual diversity into a single dominant narrative.
• Education in the AI Era: Because AI can immediately perform lower-order cognitive tasks, such as summarizing and basic programming, education must shift from testing memory to cultivating human judgment and metacognition. Baeza-Yates suggests pedagogical changes such as requiring students to actively critique and debug AI-generated answers or bringing back oral examinations.
Ultimately, the article serves as a warning that AI should be used to amplify human intellect, not atrophy it; the greatest danger is that humanity might gradually surrender the very cognitive capacities that define its existence.
As AI companies begin discussing slower development and stronger codes of conduct, a deeper question remains: Who gets to decide? This article examines how power, responsibility, and public participation should shape the rules governing A—and why these questions belong in the technology ethics classroom.
Read More
For most of human history, the hard part was making enough food. Then we got very good at it, and spent the next fifty years discovering that abundance has its own diseases. We have just done the same thing to text: generative AI made producing it close to free, while our capacity to read it has not moved an inch. I don’t think the answer is to write less out of guilt, but somebody has to be on the reading side of this, and right now the incentives are paying all of us to stand on the other one.
Read More
We are living in the early stages of an Information Shock caused by generative AI. Information Shocks are rare, but historically transformative. This one will require a reinvention of information science itself.
Read More
Three days. That’s how long Fable 5 lasted before the U.S. government ordered Anthropic to switch it off worldwide, citing a vaguely described “jailbreak” and an export control directive broad enough to sweep in Anthropic’s own employees abroad. But the recall is only half the story: Anthropic had also moved fast, pushing its most capable model to the public within months of keeping its predecessor restricted to vetted partners. From Gemini’s image generator to Tay to GPT-4o’s sycophancy rollback, this is a pattern we keep repeating, and the people who pay for it are never the ones who decided to ship.
Read More
Written communication promotes understanding, conveys authority, and inspires action. In organizations, emails often serve as a proxy for leadership presence by setting a tone, articulating expectations, and reinforcing organizational culture. Increasingly, however, this communication is no longer written solely by the sender. With a simple prompt, generative AI can instantly produce a confident response before we even have time to think or take a breath. These tools are now being suggested, and in some cases required, for drafting organizational communication. For the sender, AI-assisted writing can be helpful and even a source of relief when balancing tasks. For the recipient, AI-assisted writing may appear professional and well written. Over time, however, recipients may learn to recognize the patterns and tone of AI-generated communication. This recognition may change how people interpret the intent and authenticity of written communication.
Read More
In the past year, the focus of AI in education has shifted from generating content to evaluating its quality. While large language models can now produce vast amounts of material in seconds, ensuring that this content is accurate, reliable, and pedagogically sound remains a challenge. Emerging research shows that using AI as an evaluator is still unreliable, making human judgment more essential than ever. In this new paradigm, the real bottleneck is no longer creation but validation.
Read More
For decades, English has been a lingua franca in the research community, where it has become the principal language for publishing and conferences. But when one main language is used to share information, knowledge of this language is also needed to access information. In this way, English has become linked to information privilege: people who have mastered English can access scholarly information more easily than people who are less comfortable in this language. This has ripple effects, influencing the extent to which scientists can participate fully in scholarly communication. While the problems are clear, the solutions are trickier.
Read More
As generative AI tools move into library makerspaces, they do more than draft metadata or translate text—they reflect the power structures embedded in the spaces where they are deployed. Without intentional design, we risk automating the very inequities libraries seek to dismantle.
Read More
