AI Transparency

Twitter

Facebook

Copy Link

Key Facts

Key MethodsModel Cards, Datasheets

Required ByEU AI Act

Top News

Gartner warns poor data context makes AI agents unreliable, driving up costs and risks

Gartner research reveals that AI agents without proper data context are prone to hallucinations, biases, and unreliable outcomes. The firm predicts organizations prioritizing semantics in AI-ready data will increase agentic AI accuracy by up to 80% and reduce costs by up to 60% by 2027.

Technology2 sources

Gartner warns poor data context makes AI agents unreliable, driving up costs and risks

Technology2 sources

RPCS3 updates contribution guidelines after AI slop code caused regressions in PS3 emulator

PlayStation 3 emulator RPCS3 has updated its contribution guidelines after AI-generated code submissions caused major regressions in the emulator. The unpaid development team now requires contributors to disclose AI use and take full responsibility for their code, threatening bans for those who submit untested AI slop without understanding what they're contributing.

Technology2 sources

RPCS3 updates contribution guidelines after AI slop code caused regressions in PS3 emulator

Technology2 sources

Commerce Department quietly removes AI security testing agreement with Microsoft, Google, and xAI

The U.S. Commerce Department has deleted webpage details about a deal allowing government scientists to test AI models from Microsoft, Google, and xAI for security flaws before public release. The May 5 announcement vanished without explanation, redirecting to a general page. The removal signals potential internal disagreement about how Washington should engage with frontier AI systems.

Policy5 sources

Commerce Department quietly removes AI security testing agreement with Microsoft, Google, and xAI

Policy5 sources

NotebookLM emerges as Google AI tool that actually delivers on AI productivity promises

Google's NotebookLM is gaining attention as an AI-powered note-taking assistant that stands apart from traditional AI tools. Instead of generating information from the web, it works exclusively with user-uploaded documents to provide accurate, cited responses. Users report it excels at research, learning, and information management without the AI hallucinations that plague other tools.

Technology3 sources

NotebookLM emerges as Google AI tool that actually delivers on AI productivity promises

Technology3 sources

Neverness to Everness faces backlash as Hotta Studio admits AI art use, replaces flagged assets

Hotta Studio confirmed using AI-assisted tools for background assets in Neverness to Everness after players discovered AI-generated content throughout the gacha game. VTuber Ironmouse pulled her sponsorship deal, claiming developers misled her about AI use. The studio is now reworking flagged assets, though some replacements may still be AI-generated.

Technology2 sources

Neverness to Everness faces backlash as Hotta Studio admits AI art use, replaces flagged assets

Technology2 sources

TikTok scales back AI Overviews after feature mistakes celebrity for blueberries

TikTok has pulled back its AI Overviews feature after the experimental tool generated wildly inaccurate video summaries. The AI-powered system mistook celebrity Charli D'Amelio for a collection of blueberries and described ballroom dancers as someone striking their head with a rubber chicken. The feature will now only identify products in videos rather than provide full text summaries.

Technology3 sources

TikTok scales back AI Overviews after feature mistakes celebrity for blueberries

Technology3 sources

Spencer Pratt's AI-powered mayoral campaign reveals how technology reshapes political messaging

Former reality TV star Spencer Pratt's Los Angeles mayoral campaign has gone viral through AI-generated videos created by filmmaker Charlie Curran, portraying him as a superhero battling Mayor Karen Bass. The videos have garnered millions of views and celebrity endorsements, while sparking debate about ethical AI use in politics and the need for regulatory oversight as the technology becomes increasingly accessible to campaigns.

Entertainment3 sources

Spencer Pratt's AI-powered mayoral campaign reveals how technology reshapes political messaging

Entertainment3 sources

EU Delays AI Act Enforcement by 16 Months After Industry Pressure, Bans Nudification Apps

The EU reached a provisional agreement to delay key AI Act compliance deadlines until December 2027, responding to industry concerns about overlapping regulations. The deal also introduces an outright ban on AI-generated non-consensual intimate imagery and child sexual abuse material, following global backlash over tools like xAI's Grok chatbot.

Policy9 sources

EU Delays AI Act Enforcement by 16 Months After Industry Pressure, Bans Nudification Apps

Policy9 sources

OpenAI violated Canadian privacy laws in AI model training, federal investigation finds

Philippe Dufresne, Privacy Commissioner of Canada, found OpenAI non-compliant with federal and provincial privacy laws after a joint investigation. The probe revealed the company collected vast amounts of personal information without adequate safeguards or consent for AI model training, violating Canada's Personal Information Protection and Electronic Documents Act.

Policy3 sources

OpenAI violated Canadian privacy laws in AI model training, federal investigation finds

Policy3 sources

AI deepfakes impersonate doctors to spread medical misinformation and promote fraudulent products

The American Medical Association is calling for urgent federal and state action as AI deepfakes increasingly target doctors to promote unproven supplements and medical devices. High-profile physicians like CNN's Sanjay Gupta have been impersonated in convincing fake videos, while AI chatbots falsely claim medical licenses. The crisis threatens to erode public trust in healthcare and puts patient safety at risk.

Policy2 sources

AI deepfakes impersonate doctors to spread medical misinformation and promote fraudulent products

Policy2 sources

Illinois lawmakers debate AI liability bills as OpenAI and Anthropic clash over regulations

Illinois is considering multiple AI regulations that could reshape how companies are held accountable for harm. Two competing legislative proposals have divided major AI developers, with OpenAI supporting liability limits while Anthropic pushes for stricter oversight. The debate centers on balancing innovation with user safety as AI capabilities rapidly advance.

Policy2 sources

Illinois lawmakers debate AI liability bills as OpenAI and Anthropic clash over regulations

Policy2 sources

TikTok algorithm systematically skewed to the right during 2024 US elections, new study reveals

A groundbreaking study published in Nature found that TikTok's recommendation algorithm consistently favored pro-Republican content during the 2024 US presidential campaign. Researchers from New York University Abu Dhabi analyzed over 280,000 videos using 323 bot accounts across three states, discovering that Republican-aligned bots received 11.5% more reinforcing content than Democratic counterparts. The findings raise critical questions about algorithmic accountability and platform transparency as TikTok becomes a primary news source for young voters.

Entertainment2 sources

TikTok algorithm systematically skewed to the right during 2024 US elections, new study reveals

Entertainment2 sources

Google AI Overviews now surface expert advice from Reddit and social media in search results

Google is overhauling its AI Search experience to include direct quotes from Reddit, discussion boards, and social media posts in a new Expert Advice section. The update aims to connect users with first-hand experiences but raises questions about information reliability, especially given AI Overviews' history of citing dubious sources like sarcastic Reddit comments advising people to put glue on pizza.

Technology19 sources

Google AI Overviews now surface expert advice from Reddit and social media in search results

Technology19 sources

Google preps AI Ultra Lite plan to bridge $230 pricing gap between Pro and Ultra subscriptions

Google is developing AI Ultra Lite, a new subscription tier positioned between its $20 AI Pro and $250 AI Ultra plans. Code discovered in the Gemini macOS app reveals the tier, codenamed Neon, could cost around $100 monthly. The company is also building a dedicated usage dashboard to help subscribers track their token budget and session limits in real time.

Technology4 sources

Google preps AI Ultra Lite plan to bridge $230 pricing gap between Pro and Ultra subscriptions

Technology4 sources

Google Chrome quietly installs 4GB AI model on user devices, raising privacy and consent questions

Google Chrome has been automatically downloading a 4GB Gemini Nano AI model to users' computers since 2024, often without their knowledge. The file powers on-device AI features like scam detection and writing assistance, but users are now discovering its presence and questioning why they weren't asked for permission. While the local processing offers privacy benefits, the lack of transparency has sparked backlash.

Technology31 sources

Google Chrome quietly installs 4GB AI model on user devices, raising privacy and consent questions

Technology31 sources

Prev 1…131415…38 Next