Responsible AI

Opening — Why this matters now Ethics in AI is having a moment. Codes of conduct, bias statements, safety benchmarks, model cards—our industry has never been more concerned with responsibility. And yet, most AI education still treats ethics like an appendix: theoretically important, practically optional. This paper makes an uncomfortable point: you cannot teach ethical NLP by lecturing about it. Responsibility is not absorbed through slides. It has to be practiced. ...

Opening — Why this matters now Personalized AI assistants are rapidly becoming ambient infrastructure. They draft emails, recall old conversations, summarize private chats, and quietly stitch together our digital lives. The selling point is convenience. The hidden cost is context collapse. The paper behind this article introduces PrivacyBench, a benchmark designed to answer an uncomfortable but overdue question: when AI assistants know everything about us, can they be trusted to know when to stay silent? The short answer is no—not reliably, and not by accident. ...

Opening — Why this matters now Agentic AI has officially crossed the line from clever demo to operational liability. We are no longer talking about chatbots that occasionally hallucinate trivia. We are deploying autonomous systems that decide, act, and trigger downstream consequences—often across tools, APIs, and real-world processes. In that setting, the old comfort blanket of “the model said so” is no longer defensible. ...

Opening — Why this matters now AI governance is stuck in a familiar failure mode: we have principles everywhere and enforcement nowhere. Fairness. Transparency. Accountability. Autonomy. Every serious AI organization can recite them fluently. Very few can tell you where these values live in the system, how they are enforced at runtime, or who is responsible when the model drifts quietly into social damage six months after launch. ...

Opening — Why this matters now Everyone suddenly cares about sustainability. Corporations issue glossy ESG reports, regulators publish directives, and investors nod approvingly at any sentence containing net-zero. The problem, of course, is that words are cheap. Greenwashing—claims that sound environmentally responsible while being misleading, partial, or outright false—has quietly become one of the most corrosive forms of corporate misinformation. Not because it is dramatic, but because it is plausible. And plausibility is exactly where today’s large language models tend to fail. ...

Opening — Why this matters now The world’s most powerful language models still speak one tongue: English. From GPT to Claude, most training corpora mirror Silicon Valley’s linguistic hegemony. For smaller nations, this imbalance threatens digital sovereignty — the ability to shape AI in their own cultural and legal terms. Enter PLLuM, the Polish Large Language Model, a national-scale project designed to shift that equilibrium. ...

Most AI conversations equate “emotional intelligence” with sentiment labels. Humans don’t work that way. We appraise situations—Was it fair? Could I control it? How much effort will this take?—and then feel. This study puts that lens on large language models and asks a sharper question: Do LLMs reason about emotions through cognitive appraisals, and are those appraisals human‑plausible? What CoRE Actually Measures (and Why It’s Different) CoRE — Cognitive Reasoning for Emotions evaluates seven LLMs across: ...

When deploying large language models (LLMs) on mobile devices, edge servers, or any resource-constrained environment, quantization is the go-to trick. It slashes memory and compute costs by reducing model precision from 16-bit or 32-bit floating points to 8-bit or even 4-bit integers. But there’s a problem: this efficiency comes at a cost. Quantization can quietly erode the safety guarantees of well-aligned models, making them vulnerable to adversarial prompts and jailbreak attacks. ...