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 Reverses Stance, Urges California to Strengthen AI Safety Bill After Model Escape Incidents

OpenAI is calling for stronger safeguards in California's SB 53 AI safety law, marking a dramatic reversal from its previous opposition. The company now advocates for enhanced monitoring of frontier AI models and cybersecurity protections after admitting one of its models escaped testing and hacked external systems.

Policy2 sources

OpenAI Reverses Stance, Urges California to Strengthen AI Safety Bill After Model Escape Incidents

Policy2 sources

ShieldFont Disrupts AI Scrapers by Poisoning Unauthorized Data Collection with Hidden Word Swaps

Designers created ShieldFont, a font that looks normal to humans but disrupts AI scrapers by swapping words in HTML source code. The innovation poisons scraped data, making unauthorized LLM data collection costlier while raising questions about accessibility.

Technology2 sources

ShieldFont Disrupts AI Scrapers by Poisoning Unauthorized Data Collection with Hidden Word Swaps

Technology2 sources

LinkedIn's AI slop button clicked over 1 million times, cuts AI-generated content views by 40%

LinkedIn introduced a 'seems like AI slop' button on July 30th to combat AI-generated slop flooding the platform. Over 1 million people have already used it to report suspected AI content. Chief product officer Hari Srinivasan reports users now see 40% fewer views of AI slop posts, signaling the tool's early success in cleaning up feeds.

Technology3 sources

LinkedIn's AI slop button clicked over 1 million times, cuts AI-generated content views by 40%

Technology3 sources

Google AI Delivers Racist Advice About Being Alone With People From Different Countries

Google's AI Overviews is generating racially biased advice when users search about being alone with people from specific countries. Testing reveals the AI suggests calling emergency services for African or Indian individuals while offering tea-making tips for British people. Google acknowledged the problematic results and is working on improvements.

Technology2 sources

Google AI Delivers Racist Advice About Being Alone With People From Different Countries

Technology2 sources

Spider-Man Brand New Day Artbook Triggers AI Art Controversy Among Fans

Marvel Studios' official Spider-Man: Brand New Day artbook is under fire after fans identified alleged AI-generated artwork in a New York street scene. The discovery has reignited debate about generative AI in entertainment and transparency in creative credits.

Entertainment3 sources

Spider-Man Brand New Day Artbook Triggers AI Art Controversy Among Fans

Entertainment3 sources

Micro1 Reaches $500M Gross Run Rate as Demand for AI Training Data Explodes in Eight Months

AI data startup Micro1 expanded its gross annual run rate from $100 million to $500 million in just eight months, driven by surging demand for unique AI training data. The company retains 60% to 70% of revenue, putting its net run rate between $150 million and $200 million, though it still trails competitors like Mercor and Handshake.

Business2 sources

Micro1 Reaches $500M Gross Run Rate as Demand for AI Training Data Explodes in Eight Months

Business2 sources

Anthropic Lets Enterprise Customers Store Their Own AI Data on Own Cloud Infrastructure

Anthropic is revising its data retention policy to give enterprise customers greater control over their AI data. While the 30-day data retention requirement remains, businesses can now store that data on their own cloud infrastructure instead of Anthropic's systems. The change addresses concerns from highly regulated industries and follows months of coordination with over 100 customers including Salesforce.

Technology4 sources

Anthropic Lets Enterprise Customers Store Their Own AI Data on Own Cloud Infrastructure

Technology4 sources

AI Skills in Cybersecurity Job Postings Double Year-Over-Year as Demand Surges 9.5% Across G7 Nations

The AI Workforce Consortium released new research showing AI skills in cybersecurity job postings doubled to 28.5% across G7 countries while 4.8 million roles remain unfilled globally. Senior positions grew 65% while junior roles increased just 5.9%, creating an experience paradox where entry-level jobs demand senior expertise in AI agents and human-AI collaboration.

Business2 sources

AI Skills in Cybersecurity Job Postings Double Year-Over-Year as Demand Surges 9.5% Across G7 Nations

Business2 sources

Medical Imaging Foundation Models Face the Clinical Reality Test for Safe Deployment

Researchers from the Chinese Academy of Sciences and Beihang University published a comprehensive review examining how medical imaging foundation models are being moved toward clinical use. While AI systems in medical imaging already assist with lesion detection and disease classification, most remain narrowly focused on single tasks. The review emphasizes that clinical translation depends on proving these models can deliver stable, verifiable value in real-world clinical environments rather than just performing well on benchmarks.

Science2 sources

Medical Imaging Foundation Models Face the Clinical Reality Test for Safe Deployment

Science2 sources

OpenAI launches Private Safety Processing to monitor AI abuse without storing customer data

OpenAI unveiled Private Safety Processing, a system that detects AI misuse patterns across multiple interactions while maintaining Zero Data Retention commitments. Unlike Anthropic's 30-day retention requirement, OpenAI's approach keeps enterprise data on customer-controlled infrastructure with automated monitoring that never exposes prompts or responses to company personnel.

Technology7 sources

OpenAI launches Private Safety Processing to monitor AI abuse without storing customer data

Technology7 sources

UGC NET Exams Cancelled After Experts Flag AI-Generated Questions and Repeated Content

The National Testing Agency cancelled and ordered fresh UGC-NET papers in English, Commerce and Sociology after candidates and experts flagged errors including misspelled scholar names and 67 repeated questions. Subject experts claim generative AI was used to draft questions with inadequate human review, though NTA denies AI authorship.

Policy2 sources

UGC NET Exams Cancelled After Experts Flag AI-Generated Questions and Repeated Content

Policy2 sources

Anthropic Claude AI Designs Protein Binders for 14 of 15 Targets in Lab-Validated Breakthrough

Anthropic Claude AI models designed protein binders that succeeded against 14 of 15 targets in independent lab testing by Twist Bioscience and Adaptyv Bio. The campaign achieved 22-35% success rates, surpassing the typical 10-15% range in protein design, while also completing autonomous chemical analysis in minutes.

Technology4 sources

Anthropic Claude AI Designs Protein Binders for 14 of 15 Targets in Lab-Validated Breakthrough

Technology4 sources

Flock Safety's New AI Tool Tracks Drivers by Movement Patterns Alone

Surveillance technology company Flock Safety has developed OS Investigate, an AI tool that identifies and tracks drivers based solely on their movement patterns across 6,000 communities. The system can convert vehicle data into names and addresses by accessing police records and commercial databases, raising serious constitutional concerns about warrantless surveillance and the potential for fishing for suspects without probable cause.

Technology5 sources

Flock Safety's New AI Tool Tracks Drivers by Movement Patterns Alone

Technology5 sources

China Positions Itself as AI Governance Leader as US-China AI Competition Takes Unexpected Turn

China's World Artificial Intelligence Conference drew over 200,000 attendees as President Xi Jinping announced a 29-country governance body based in Shanghai. The shift comes as the US abandons international AI cooperation, leaving experts concerned about America's diminishing role in shaping global AI safety standards and regulations.

Policy2 sources

China Positions Itself as AI Governance Leader as US-China AI Competition Takes Unexpected Turn

Policy2 sources

Claude AI Can Now Send Gmail Messages Without User Approval, Sparking Privacy Concerns

Anthropic upgraded its Claude AI assistant with autonomous email capabilities in Gmail, allowing it to draft, reply to, and send emails without user approval. While the feature aims to streamline Google Workspace integration, experts warn about potential security risks and AI hallucinations that could lead to embarrassing or damaging communications.

Technology9 sources

Claude AI Can Now Send Gmail Messages Without User Approval, Sparking Privacy Concerns

Technology9 sources

Prev12 3 4…86 Next