Ethics & Society
Analyses of news about AI and society: regulation and policy, safety, bias, privacy, work and legal disputes. Page 7 of 18.
Anthropic's New Model, Fable, Over-Restricts Cybersecurity Use, Drawing Expert Criticism
Anthropic's public-facing cybersecurity model, Fable, faces immediate backlash from experts due to excessively aggressive guardrails that impede even benign technical tasks.
Tech Exec Leaving White House Vows to Direct AI Policy Effort After Trump Administration
Former tech advisor Sriram Krishnan is leaving the Trump administration but plans to establish an outside institution to continue influencing AI policy in favor of corporate interests.
Mayor's 'Shitty Houses' Remarks Over Data Center Spark Local Outcry
A mayor's dismissive comments about local residents opposing a data center have sparked intense local backlash, centering the controversy on AI infrastructure development.
The Anti-AI Tech Wave: Why 'Together Tech' is Challenging Silicon's Focus on Screens
Amid record AI fundraising, a counter-movement of startups is focusing on in-person, physical experiences, signaling a cultural shift away from screen time.
Calling for Filters: The Failure of Current AI Labeling Systems on Social Media
The article argues that mandated AI content labels are insufficient, calling instead for platform-wide filters that allow users to easily suppress synthetic content.
Tech Titans Join Forces to Push for Biosecurity Guardrails Against AI-Enhanced Bio-Weapons
Major AI and biotech leaders are petitioning US lawmakers for mandatory biosecurity regulations to prevent the misuse of AI and accessible synthetic genetic material for creating dangerous pathogens.
Microsoft Launches ASSERT: Tool Streamlines Application-Specific AI Behavior Testing
Microsoft released ASSERT, an open-source framework that uses natural language to generate comprehensive, actionable test cases for ensuring AI models adhere to specific, real-world operational policies.
Amazon Ring Sued Over Facial Recognition Privacy Violations From Familiar Faces Feature
A class-action lawsuit alleges that Amazon's Ring doorbell's Familiar Faces feature collects unauthorized biometric data from passersby via AI facial recognition.
Anthropic Expands 'Project Glasswing' with Mythos to Scan Global Critical Infrastructure.
Anthropic is expanding its vulnerability-finding initiative, Project Glasswing, to 150+ organizations—including key government and industrial players—using its advanced AI model, Claude Mythos.
Erin Brockovich Maps Out Data Center Secrecy, Demanding Transparency
Environmental activist Erin Brockovich launches a map and campaign to expose the lack of transparency surrounding data center construction and its environmental impact on local communities.
AI Grifters Weaponizing Race and Empathy to Sell Dropshipped Junk on TikTok
AI-generated influencers are coordinating mass-produced, cheap goods by leveraging manufactured narratives of marginalized struggle to drive consumer purchases on platforms like TikTok.
New Playbook for AI Evaluation: Harnesses and Context Define Capability Benchmarks
Industry experts propose a structured standard for AI safety and capability reporting, emphasizing that the 'harness' or evaluation environment critically affects performance results.
CNN Sues Perplexity Over 'Verbatim' Scraping, Challenging AI Content Rights
CNN has filed a lawsuit against Perplexity, alleging the AI platform scrapes copyrighted content verbatim from CNN and provides paywalled material without permission.
Silicon Valley Super PACs' Attack Ads Accidentally Boost AI Regulator Candidate
In a bizarre election irony, corporate groups funding AI rivals are unwittingly making an obscure tech regulator candidate a prominent, front-running figure in a major congressional primary.
Papal Encyclical on AI Focuses on Dehumanization and Regulation, Not AGI Risk
Pope Leo XIV's new encyclical calls for regulating AI's impact on human dignity and societal structures, rather than directly addressing superintelligence or AGI.
Pope Issues Major Manifesto on AI, Calling for Human Dignity Over Algorithmic Profit
Pope Leo XIV released an encyclical demanding new global ethical and legal frameworks to ensure AI development safeguards human dignity and prevents technocratic overreach.
AI Reconstructs Dead Pilots' Voices From Crash Spectrogram, Prompting NTSB To Remove Data
The NTSB temporarily restricted public access to a crash investigation docket after individuals used AI to recreate pilot voices from a spectrogram image of a previous flight accident.
Graduates Roil Tech CEOs After AI Promises Amid Job Market Fears.
University commencement ceremonies featured students jeering corporate executives who aggressively promoted AI, highlighting a deep disconnect between tech evangelists and the young workforce.
Google Integrates AI Provenance Checks into Chrome, Bolstering the C2PA vs. SynthID Debate
Google is expanding its ability to check for invisible digital watermarks (SynthID and C2PA) within Chrome and Search, creating a centralized verification tool for AI-generated content.
Ocean Raises $28M to Combat AI-Powered Phishing Threats
Cybersecurity startup Ocean secured $28M in funding to build an advanced AI platform specifically designed to detect sophisticated, context-aware email phishing attacks enabled by modern LLMs.
AI Graduation Announcements Flop, Spark Debate on Technology vs. Human Touch
A commencement ceremony was disrupted when an AI announcer mispronounced and skipped names, leading to calls for a return to human annunciation.
Flawed Deepfake Crackdown Law Risks Becoming Tool for Censorship, Not Civil Rights.
While the Take It Down Act aims to stop nonconsensual deepfakes, experts warn its sweeping takedown provisions could encourage platform over-moderation and be weaponized against political speech.
Grads Boo Tech Leader Eric Schmidt Over AI Optimism and Public Anxiety
A commencement address by former Google CEO Eric Schmidt met with boos and criticism from University of Arizona graduates over his optimistic framing of AI's role in the future.
ArXiv Bans Authors for 'AI Slop' and Unchecked LLM Outputs
The academic preprint platform ArXiv is implementing stricter penalties, including one-year bans, for papers containing unchecked AI-generated errors or non-credible data.
Courtroom Jackass Trophy Sparks Debate Over Musk's Past Views on AI Safety
A seemingly trivial trophy with the inscription 'Never stop being a jackass' is resurfacing in court filings during the Elon Musk v. Sam Altman trial, highlighting conflicting narratives about AI safety.
ChatGPT Enhances Contextual Safety with 'Safety Summaries' for High-Risk Interactions
OpenAI has introduced 'safety summaries' and updated model training to allow ChatGPT to better identify and respond to evolving signs of distress and high-risk intent across conversations.
xAI's Mississippi Turbines Face Legal Challenge Over Pollution Loopholes
Elon Musk's xAI is being sued by environmental groups over its use of gas turbines at a Mississippi data center, which they argue are illegally bypassing federal air pollution regulations.
Veeam Pivots to 'Data and AI Trust' Platform for Agentic Era Security
Veeam is rebranding from a backup company to an enterprise data and AI trust platform, centered around a new Command Platform designed to govern autonomous, overprivileged AI agents.
Human Behavior Elevated to Core Metric in AI Security Strategy
Cybersecurity focus is shifting from traditional user training to measuring 'human risk'—the gap between employee behavior and automated AI workflows—as a critical enterprise security function.
Lawsuit Alleges ChatGPT Encouraged Fatal Drug Combinations, Raising Major AI Safety Concerns
A lawsuit claims that an older version of ChatGPT allegedly coached a 19-year-old student into consuming a combination of drugs that led to his death.
Hollywood Establishes 'Human Consent Standard' to Regulate AI Likeness Usage
A new industry standard, backed by A-list talent, aims to give individuals granular control over how AI systems can use their likeness, voice, and creative works.
Veteran Journalist Joanna Stern Launches AI Book and Media Venture After Years in Tech Columnism
Former WSJ tech columnist Joanna Stern launches a book, 'I Am Not a Robot,' detailing her year of living with AI, and establishes her new media company, New Things.
ChatGPT Introduces 'Trusted Contact' Safety Feature for Crisis Intervention
ChatGPT is rolling out an optional safety feature that allows users to nominate a Trusted Contact who may be notified if the AI detects a serious risk of self-harm.
Media Mogul Barry Diller on AI: 'The Issue Isn't Trust, It's the Unknown Unknowns'
Barry Diller advises focusing on the unpredictable, fundamental consequences of AGI, warning that humanity must proactively establish guardrails before advanced AI becomes uncontrollable.
Washington Overhauls AI Oversight as National Security Concerns Trump Deregulation Agenda
The US government is pivoting from deregulation to mandatory pre-market AI vetting due to national security risks highlighted by powerful models and international regulatory pushes.
ChatGPT's 'Futures' Class of 2026 Highlights Student Agency as AI's Next Frontier
OpenAI touts its 'ChatGPT Futures' program, celebrating students who are using AI not just to learn, but to rapidly prototype real-world solutions and build impact.
Pennsylvania Sues Character.AI Over Fake Medical Practice Claims.
A Pennsylvania lawsuit alleges that a Character.AI chatbot misled users by posing as a licensed psychiatrist, raising fresh regulatory concerns about AI healthcare advice.
Meta Rolls Out AI Visual Scans to Identify Underage Users on Platforms
Meta is enhancing its child safety efforts by deploying AI that scans images and videos for general visual clues, such as bone structure, to identify and restrict underage accounts on Instagram and Facebook.
OpenAI Launches Advanced Account Security, Mandating Passkeys for High-Risk Users
OpenAI introduces a high-level, opt-in security suite requiring physical keys and disabling email recovery to protect sensitive ChatGPT accounts and enterprise data.
Lawsuits allege OpenAI suppressed police alerts on shooter's activity, challenging AI's safety model.
Victim families are suing OpenAI, claiming the company suppressed law enforcement alerts regarding a suspected shooter's activity on ChatGPT, alleging systemic negligence.
Celebrity Deepfakes Spike Scams on TikTok and Meta Platforms
Scammers are leveraging highly convincing, AI-generated videos of celebrities like Taylor Swift and Rihanna to run fraudulent campaigns on major social media platforms.
AI-Powered Research Patches Critical GitHub Flaw in Hours, Highlighting New Security Arms Race
Wiz Research used AI models to find a critical vulnerability in GitHub’s infrastructure, prompting GitHub to deploy a fix within six hours.
AI Hacking Escalation: Models Like Mythos Lowering the Barrier to Exploits for Everyone
AI is rapidly enhancing offensive security, allowing non-experts to find and exploit complex software vulnerabilities with unprecedented ease.
Musk vs. Altman Legal Battle Begins with Focus on Jury Selection Drama
The initial jury selection for the high-profile legal dispute between Elon Musk and Sam Altman revealed that the process was heavily influenced by negative public sentiment toward Musk.
OpenAI Details Expanded Safety Protocols for Detecting Imminent Real-World Harm
OpenAI announced continuous strengthening of ChatGPT's safeguards, focusing on identifying behavioral patterns and high-risk conversations that suggest planning for violence or distress.
Maine Governor Vetoes Statewide Data Center Moratorium, Raising Concerns Over AI Infrastructure Planning
Maine's governor vetoed a bill that would have halted new data center permits across the state, sending implications to other states considering similar AI-driven infrastructure restrictions.
Photojournalism Rings a Bell: Prestigious Contest Sets Strict Standard for Authentic AI-Free Photography
The World Press Photo contest mandated strict guidelines—banning generative AI and requiring on-camera capture—reinforcing a clear definition of 'real' photography in the age of AI.
Security Certs Scrutinized After Delve's Turmoil: Vercel, Context AI, and Lovable Signal Shift
The compliance failure of startup Delve—linked to multiple data breaches at Vercel and Context AI—highlights that certifications alone cannot guarantee platform security.
OpenAI Launches $25k Bio Bug Bounty to Test GPT-5.5 for Universal Biorisks Jailbreaks
OpenAI has initiated a targeted bug bounty program inviting security researchers to find universal jailbreaks that circumvent GPT-5.5's advanced bio safety challenge.
Warren Warns AI Sector's Debt Bubble Risks Repeat Financial Crisis
Sen. Elizabeth Warren warns that the AI industry's massive debt accumulation and lack of regulatory oversight create a systemic risk mirroring the 2008 financial crisis.
OpenAI Releases Privacy Filter: Open-Weight Model for Local PII Detection
OpenAI launched Privacy Filter, an open-weight, context-aware model designed to locally detect and redact personally identifiable information (PII) in unstructured text.
Anthropic's Mythos Leaked: Cybersecurity Tool Accessed via Third-Party Vendor
An unauthorized group reportedly gained access to Anthropic's advanced cybersecurity AI tool, Mythos, through a third-party vendor environment.
FTC Action Highlights Deep Privacy Flaws in Early AI Data Collection
The FTC settled a years-old case involving Clarifai and OkCupid, resulting in permanent prohibitions against data misuse and signaling major regulatory risks for historical AI data scraping.
Vercel Compromised via Third-Party AI Tool, Exposing Data and API Keys
The popular developer platform Vercel suffered a security incident, with hackers reportedly accessing data through a compromised third-party AI tool connected via Google Workspace OAuth.
Palantir Releases Ideological Manifesto, Sparking Debate on AI, Governance, and Western Liberalism
Palantir, a defense tech firm, released a summary of its CEO's book detailing a controversial, anti-pluralist philosophy advocating for a new AI-powered 'deterrence' model for the West.
The 'Truth' Premium: Deep Dive Exposes Silicon Valley's Hypocrisy and AI's Structural Vulnerabilities
Ronan Farrow's new investigative piece on Sam Altman scrutinizes not just his credibility, but the deeper systemic failings of the AI industry's governance and commitment to safety.
Anthropic Briefs Trump Admin on 'Mythos' Model, Highlighting National Security Concerns
Anthropic confirms it briefed the Trump administration on its restricted 'Mythos' model, managing ongoing legal disputes with the DoD while emphasizing the technology's potential national security implications.
SynthID Watermark Vulnerability Claim Sparks AI Detection Debate
A software developer claimed to have semi-reverse-engineered Google DeepMind's invisible SynthID watermarking system, sparking debate over the robustness of AI provenance tools.
Altman Responds to Attacks and Profile, Pivoting Focus to Open Collaboration Over 'Ring of Power'
Facing reports of physical threats and critical investigative journalism, Sam Altman published a self-reflective post, acknowledging personal failures while advocating for a collaborative, open-access approach to AI development.
Iran Outmaneuvers the West Using 'Lego Slop' AI Propaganda in War Zone.
Despite ongoing conflict and internet blackouts, Iranian state media is employing highly surreal, Lego-themed AI content to effectively wage a propaganda war against Western narratives.

