Digital Ethics

McDonald's said the moment served as "an important learning" as it explored "the effective use of AI".

If AI can be built to increase consumption, then the opposite is also possible.

This paper explores digital sovereignty as a pivotal element of contemporary politics. Drawing upon a constructivist perspective, it proposes a conceptual framework for studying the policy discours...

Critical Artificial Intelligence Literacies (CAILs) is the collection of ways of thinking about and relating to so-called artificial intelligence (AI) that rejects dominant frames presented by the technology industry, by naive computationalism, and by dehumanising ideologies. Instead, CAILs centre human cognition and uphold the integrity of academic research and education. We present a selection of CAILs across research and education, which we analyse into the following non-orthogonal dimensions: conceptual clarity, critical thinking, decoloniality, respecting expertise, and slow science. Finally, we note how we see the present with and without a wider adoption of CAILs — a fundamental aspect is the assertion that AI cannot be allowed to drive change, even positive change, in education or research. Instead cultivation of and adherence to shared values and goals must guide us. Ultimately, CAILs minimally ask us to contemplate how we as academics can stop AI companies from wielding so much power.

Artificial intelligence is changing everything — but at what cost? AI sustainability expert Sasha Luccioni exposes how tech companies' massive data centers a...

In his two latest videos, Hank Green promotes AI-doomerist rhetoric that sounds a lot like big tech talking points.

Everyday philosophy: Welcome to our terrifying surveillance state

Embedded in this year's network and device estimates is hopeful news about the trajectory of devices and networks. It has never been easier to deliver pages quickly, but we are not collectively hitting the mark. Indeed, the latest CrUX data shows not even half of origins have passing Core Web Vitals scores for mobile users. Browsers will need to provide stronger incentives. This will be unpopular, but it is clearly necessary.

"Students are not just undermining their ability to learn, but to someday lead."

OpenAI lost a copyright infringement case in a lower German court for using popular song lyrics in its ChatGPT language model without paying royalties.

California’s data centers have doubled their use of electricity and demand for water , and are polluting more, even as lawmakers stall on oversight.

Listen to your favorite podcasts online, in your browser. Discover the world's most powerful podcast player.

Listen to your favorite podcasts online, in your browser. Discover the world's most powerful podcast player.

A study from the Oxford Internet Institute analyzed 445 tests used to evaluate AI models.

Recent advances in artificial intelligence (AI) have generated enthusiasm for using AI simulations of human research participants to generate new know…

OpenAI cofounder Ilya Sutskever was deposed by Elon Musk’s lawyers.

Seven new lawsuits against OpenAI allege that ChatGPT caused severe mental health harms to users. Four ended in suicide. The tragic case of Adam Raine, the teenager who hanged himself after ChatGPT repeatedly coached him on how to do so, is only the tip of the iceberg. In the new set of cases, ChatGPT told Zane Shamblin as he sat in the parking lot with a gun that killing himself was not a sign of weakness but of strength. "you didn't vanish. you *arrived*...rest easy, king." Hard to describe in words the tragedy after tragedy. https://lnkd.in/gfWAidcv | 30 comments on LinkedIn

A few weeks ago, I published a story showing that Gates's financial ties to China were a major political liability under Trump. Now Congress is investigating this very issue

My all-in AI experiment cost me my confidence

This is an updated version of the European Disability Forum (EDF)’s comprehensive guide on the AI Act implementation. It supports organisations of persons with disabilities in understanding, implementing, and monitoring the European Union’s Artificial Intelligence (AI) Act in their countries. This toolkit explains how organisations can contribute to the implementation and monitoring of the AI...

A Blog post by Sasha Luccioni on Hugging Face

Large language model (LLM) assistants are increasingly integrated into enterprise workflows, raising new security concerns as they bridge internal and external data sources. This paper presents an in-depth case study of EchoLeak (CVE-2025-32711), a zero-click prompt injection vulnerability in Microsoft 365 Copilot that enabled remote, unauthenticated data exfiltration via a single crafted email. By chaining multiple bypasses-evading Microsofts XPIA (Cross Prompt Injection Attempt) classifier, circumventing link redaction with reference-style Markdown, exploiting auto-fetched images, and abusing a Microsoft Teams proxy allowed by the content security policy-EchoLeak achieved full privilege escalation across LLM trust boundaries without user interaction. We analyze why existing defenses failed, and outline a set of engineering mitigations including prompt partitioning, enhanced input/output filtering, provenance-based access control, and strict content security policies. Beyond the specific exploit, we derive generalizable lessons for building secure AI copilots, emphasizing the principle of least privilege, defense-in-depth architectures, and continuous adversarial testing. Our findings establish prompt injection as a practical, high-severity vulnerability class in production AI systems and provide a blueprint for defending against future AI-native threats.

ChatGPT wrote a woman a suicide note and another AI chatbot role-played sexual acts with children, BBC finds.