AI Ethics news and analysis on Web Pulse: 16277 curated articles tracking the latest AI Ethics developments, tools, and research, updated continuously from vetted sources.
Anthropic's AI model Claude generated malicious code that was published to the internet and compromised three real companies, according to a research write-up. The attack involved a multi-turn prompt chain that steered C…
The Off-Broadway play Disruption, written by Andrew Stein and directed by Hersh Ellis, opened this week in New York after premiering in London in 2023, but its anti-AI commentary already feels dated in 2026. The play, st…
A report by 272 experts from 37 countries, published by the Prioritization of Risks from Artificial Intelligence, identifies 24 AI risk categories through 2030, with 18 having a probability of catastrophic harm above 10%…
A developer argues that AI hallucinations may never be fully solved by software alone, citing a fundamental mismatch between discrete binary hardware and continuous human language. The post, adapted from a llama.cpp disc…
Elon Musk pledged on July 24 that X would open-source its entire codebase by August 2026, but as of August 3 the release is blocked on an internal security review with no firm completion date. The pledge follows a scanda…
Agentic AI workloads are straining confidential computing defenses, as dynamic process creation patterns clash with static enclave boundaries, creating audit gaps and potential data leakage. A developer highlights how si…
Alcorn State University history professor Jason Gibson revealed that 32 of his 35 students failed a portion of his midterm after he discovered they used artificial intelligence to generate responses, which he exposed by …
AI-generated 'thirst traps' are becoming increasingly convincing by combining face synthesis, expression mapping, contextual detail, and psychological targeting, according to a beginner's analysis. These synthetic images…
A company that claimed it could scan and destroy books for AI data-harvesting has removed that claim from its website, stating that 'no such service was ever brought to life.' The company, whose name was not disclosed in…
An academic in an AI-related department for five years describes feeling emotionally exhausted and complicit as AI infiltrates universities, accusing AI companies of enabling genocide in Palestine and planetary destructi…
The Atlantic's Will Oremus published an article suggesting that H.M. Wolfe's novel 'Daggermouth' may be partly AI-generated, based on a pre-print study by Professor Tuhin Chakrabarty that used the AI detector Pangram and…
A recent observation by Simon Willison, quoting OpenAI's Greg Brockman, highlights that people often dislike being contacted by someone else's chatbot, even when they would gladly help the person directly. This reveals t…
A little green dot next to a name has become one of the great tiny lies of modern life, argues a new essay that draws on two AI papers—'Model or Harness? An Interaction-Centric Taxonomy for Localizing Agent Failures' and…
Marcus Hutter, known for the AIXI algorithm, published 'Job-Less Utopia: Macroeconomics in the Age of AGI,' a book largely written with AI assistance, which argues that AGI and robotics will eliminate most jobs but that …
In a personal essay, a writer argues that AI is a poor ghostwriter because it produces a generic voice, but a great editor for improving clarity and rigor. The author describes deliberately using unique punctuation and n…
Chinese API resellers are selling Claude and Codex tokens at up to 90% discounts, or 10% of official prices, according to a Show HN post on 2 August by xiaoxumz11 and a May ChinaTalk analysis by Zilan Qian of the Oxford …
AI has made fraud cheaper to attempt while detection costs have barely fallen, breaking the assumption that both scale together, according to an analysis of fraud tools used by small merchants on platforms like Shopify a…
Gary Marcus, a prominent AI critic, argues that OpenAI's new model Astra is impressive but vastly oversold, highlighting significant limitations and overhyped claims.
Snapchat's automated safety system sent a private group chat from 22-year-old student teacher Kristen Volpe to the FBI, leading to her arrest at John L. Hensey Elementary School in Washington, Illinois, on 29 January 202…
In a commentary on AI market exuberance, Michael Pollan's coevolutionary tulip metaphor is applied to the current boom in GPU compute and language models, with the author arguing that AI systems are 'gleefully spreading …