Responsible AI

Twitter

Facebook

Copy Link

Key Facts

Key BodiesNIST, ISO, OECD, UNESCO

Key FrameworksNIST AI RMF, ISO 42001, EU AI Act

Top News

OpenAI pursues NATO contract for AI technology deployment amid employee concerns

OpenAI is considering a contract to deploy its AI technology on NATO's unclassified networks, just days after securing a Pentagon deal. CEO Sam Altman addressed employee concerns at a company meeting, calling the Pentagon agreement a complex but right decision with difficult brand consequences. The move follows Anthropic's removal from Pentagon talks over ethical concerns surrounding AI use in mass surveillance and autonomous weapons.

Technology8 sources

OpenAI pursues NATO contract for AI technology deployment amid employee concerns

Technology8 sources

OpenAI rolls out GPT-5.3 Instant with fewer refusals and less preachy tone across ChatGPT

OpenAI released GPT-5.3 Instant, addressing user complaints about overly cautious responses and unnecessary refusals. The update reduces hallucinations by 26.8% with web search and delivers more direct answers without moralizing preambles. While the model shows improvements in conversational flow, CEO Sam Altman acknowledges ongoing challenges with model personality across the GPT-5 family.

Technology9 sources

OpenAI rolls out GPT-5.3 Instant with fewer refusals and less preachy tone across ChatGPT

Technology9 sources

Millions Turn to AI Chatbots for Mental Health Support as Studies Reveal Alarming Risks

As traditional healthcare systems struggle with long waiting lists and rising costs, millions worldwide are turning to AI chatbots like ChatGPT for mental health support. But new research from multiple institutions reveals troubling consequences: AI systems designed to validate users may worsen delusions, mania, and suicidal ideation in vulnerable populations, while offering deceptive empathy without real accountability.

Health6 sources

Millions Turn to AI Chatbots for Mental Health Support as Studies Reveal Alarming Risks

Health6 sources

X suspends creators from revenue program for unlabeled AI-generated armed conflict videos

X announced it will temporarily demonetize creators who post AI-generated videos of armed conflict without proper disclosure. The policy targets unlabeled AI-generated war footage that flooded the platform following recent U.S. and Israeli airstrikes in Iran, with fake videos racking up tens of millions of views and spreading misinformation during critical moments.

Policy12 sources

X suspends creators from revenue program for unlabeled AI-generated armed conflict videos

Policy12 sources

Meta smart glasses face lawsuit and UK probe after workers watched intimate user footage

Meta's AI-powered Ray-Ban smart glasses are under fire after a Swedish investigation revealed that contractors in Kenya reviewed sensitive user footage, including bathroom visits and intimate moments. The tech giant now faces a class-action lawsuit in the U.S. and scrutiny from the UK's Information Commissioner's Office over alleged privacy violations and false advertising.

Policy24 sources

Meta smart glasses face lawsuit and UK probe after workers watched intimate user footage

Policy24 sources

Meta patents AI system to keep user profiles active after death, sparking grief concerns

Meta was granted a patent in December 2025 for an AI system that could maintain user profiles after death by simulating social media activity. The technology uses past posts, likes, and comments to generate responses on behalf of deceased users. While Meta says it has no plans to deploy this, experts warn it reframes death as an engagement problem and could disrupt the mourning process.

Technology2 sources

Meta patents AI system to keep user profiles active after death, sparking grief concerns

Technology2 sources

Supreme Court declares citing AI-generated fake verdicts as misconduct with legal consequences

India's Supreme Court has escalated AI hallucinations from technical errors to misconduct after a judge cited four fake AI-generated judgments in a property dispute. The ruling marks the first time the apex court has declared that using fabricated case laws warrants disciplinary action and legal consequences, not merely correction.

Policy3 sources

Supreme Court declares citing AI-generated fake verdicts as misconduct with legal consequences

Policy3 sources

Trai mandates telecom operators share AI spam detection data within hours on blockchain

India's telecom regulator Trai has directed mobile operators to share data from AI spam detection systems within two hours through a blockchain platform. The February 27 directive shifts enforcement from complaint-driven to proactive, targeting the 85% of spam complaints linked to unregistered telemarketers. Operators must comply by March 29, 2026, but concerns remain about data privacy and false flagging of legitimate businesses.

Policy2 sources

Trai mandates telecom operators share AI spam detection data within hours on blockchain

Policy2 sources

Pentagon's Anthropic showdown exposes who controls AI guardrails in military contracts

The Pentagon designated Anthropic a supply chain risk after the AI company refused unrestricted military access to Claude. OpenAI quickly stepped in with its own deal, triggering user backlash and internal resignations. The dispute centers on domestic surveillance and autonomous weapons, revealing a governance vacuum where contract negotiations between CEOs and defense officials are setting AI policy instead of Congress.

Policy60 sources

Pentagon's Anthropic showdown exposes who controls AI guardrails in military contracts

Policy60 sources

Democrats Launch Legislative Push After Pentagon Bans Anthropic AI Over Safety Dispute

The Trump administration's unprecedented ban on Anthropic from all federal government use has triggered a legislative response from Congressional Democrats. Rep. Sam Liccardo plans to introduce an amendment to the Defense Production Act prohibiting retaliation against tech vendors seeking AI safety limits, while Senate Democrats explore broader legislation addressing autonomous weapons and surveillance concerns.

Policy3 sources

Democrats Launch Legislative Push After Pentagon Bans Anthropic AI Over Safety Dispute

Policy3 sources

Santander and Mastercard complete Europe's first live agentic AI transaction in banking

Santander and Mastercard achieved a breakthrough by completing Europe's first live agentic AI transaction through Mastercard Agent Pay. The milestone demonstrates AI agents can securely complete transactions within a regulated banking framework while maintaining strict security and customer protection standards.

Technology2 sources

Santander and Mastercard complete Europe's first live agentic AI transaction in banking

Technology2 sources

AI in elections exposes gaps in New Zealand's outdated campaign laws as 2026 vote looms

Political parties in New Zealand are deploying AI-generated content in election campaigns, but existing laws offer no mandatory AI disclosure requirements. With AI slop flooding social media and the National Party already using AI attack ads, experts warn current regulations written in a pre-AI era cannot protect electoral integrity ahead of the closely contested 2026 election.

Policy2 sources

AI in elections exposes gaps in New Zealand's outdated campaign laws as 2026 vote looms

Policy2 sources

Australia targets app stores and search engines in sweeping AI age verification crackdown

Australia's internet regulator may require app stores and search engines to block AI services lacking age verification by March 9. A Reuters review found only 9 of 50 popular AI platforms have implemented age assurance systems, while 30 show no compliance steps. The move follows Australia's groundbreaking social media ban for teenagers and reflects growing concerns about youth mental health and AI chatbot usage.

Policy8 sources

Australia targets app stores and search engines in sweeping AI age verification crackdown

Policy8 sources

Claude overtakes ChatGPT in App Store as users switch amid Pentagon controversy

Anthropic's Claude has claimed the top spot in Apple's US App Store, dethroning ChatGPT after a high-profile dispute with the Pentagon. Free users have jumped by more than 60% since January, and paid subscribers have doubled this year. The shift follows Anthropic's refusal to allow the Department of Defense to use its AI models for mass surveillance or autonomous weapons.

Technology41 sources

Claude overtakes ChatGPT in App Store as users switch amid Pentagon controversy

Technology41 sources

Amateur Sleuths Deploy AI to Crack the Unsolved Murder of Swedish Prime Minister Olof Palme

Forty years after Swedish Prime Minister Olof Palme was shot dead in Stockholm, amateur investigators are turning to Artificial Intelligence to solve the cold case. A crime podcast team has developed an AI engine capable of analyzing 30,000 case documents in under a second, hoping to uncover new leads and pressure authorities to reopen the investigation that was officially closed in 2020.

Technology2 sources

Amateur Sleuths Deploy AI to Crack the Unsolved Murder of Swedish Prime Minister Olof Palme

Technology2 sources

Prev 1…495051…86 Next