Audits, standards and certifications — the operational side of making AI systems lawful and accountable.
321 articles
A new EU briefing argues that generative AI is changing tasks before eliminating jobs, shifting value toward judgment, domain expertise, quality control, and accountability. Under Article 4 of the EU AI Act, providers and deployers must take measures to ensure employees and relevant personnel have sufficient AI literacy.
虎嗅 · 11d ago
The EU’s new AI Act transparency obligations took effect on August 2, requiring companies to disclose AI interactions and identify content generated or altered by AI. The requirements vary between providers that develop AI systems and deployers that use them.
The Verge · 47d ago
The European Union has announced that new AI rules are now enforceable across the bloc. The development makes AI transparency compliance an active regulatory concern for organizations operating in the EU.
Engadget · 47d ago
The Baker v. Index Exchange ruling allows private parties to use the DOJ's 'Bulk Sensitive Data' (BSD) rule as a basis for litigation, significantly increasing compliance risks for Chinese companies using real-time bidding (RTB) advertising.
虎嗅 · 65d ago
OPPO and other major manufacturers have completed the mandatory regulatory filing for their on-device generative AI services. The company states this is a standard compliance action to align with national AI model regulations.
36氪 · 66d ago
China has issued new guidelines quantifying 'important data' thresholds for financial information services, impacting how AI companies handle large-scale user datasets. This marks a shift toward stricter, standardized data classification for AI training and inference.
虎嗅 · 75d ago
Anthropic has been caught embedding hidden identification logic in its Claude Code client to detect and block users in China. The mechanism uses local system data and steganography in system prompts to bypass detection, sparking significant backlash regarding user privacy and transparency.
虎嗅 · 78d ago
Anthropic PBC has received US government approval to expand access to its Mythos 5 AI model. The clearance follows a resolution of national security concerns raised by the Trump administration.
Bloomberg Technology · 82d ago
The US government has partially lifted export control restrictions on the Claude Mythos 5 model. Access is now limited to a specific whitelist of authorized institutions due to previous security concerns regarding jailbreaking.
cnBeta (Full RSS) · 84d ago
This article outlines the compliance requirements for the EU AI Act across different risk tiers. It provides a roadmap for engineering teams to operationalize AI governance through 2027.
Docker Blog · 84d ago
Stripe shares insights into building production-grade AI agents for financial compliance. The article covers the ReAct framework, infrastructure decisions, human oversight, and cost optimization via prompt caching.
AWS Machine Learning Blog · 85d ago
The US government is reportedly requesting access to Meta's AI models to conduct security and safety evaluations. This move reflects growing regulatory scrutiny over the potential risks associated with open-weight AI technology.
Engadget · 87d ago
The National Radio and Television Administration (NRTA) has released a draft regulation requiring all micro-dramas generated or produced using AI technology to include clear, prominent disclosure labels.
IT之家 · 87d ago
Anthropic has taken its two newest AI models offline to comply with a U.S. regulatory order.
GeekWire · 91d ago
The US Department of Commerce has issued a warning to Anthropic, requiring government approval before allowing foreign nationals to access its advanced AI models. Failure to comply could result in criminal and civil penalties.
cnBeta (Full RSS) · 94d ago
Anthropic has globally withdrawn the Claude Fable 5 model following US government export control directives. The action was triggered by security vulnerabilities that allowed potential bypasses of safety boundaries.
cnBeta (Full RSS) · 95d ago
Anthropic received a US export control directive requiring the suspension of access to its Mythos 5 and Fable 5 models for all foreign nationals. The company had to disable these products globally while attempting to negotiate with the Trump administration.
The Verge · 95d ago
The US government has directed Anthropic to restrict access to its AI models due to concerns that they could be diverted for use by foreign military intelligence. This move underscores the growing intersection of national security and generative AI deployment.
iTNews Australia · 96d ago
Starting September 11, 2026, manufacturers selling many products with digital elements in the EU must report actively exploited vulnerabilities within 24 hours of becoming aware of them. GitLab emphasizes continuous software-supply-chain detection as the practical compliance challenge.
GitLab Blog · 9d ago
The Trump administration plans to confront Australia over proposed laws requiring digital platforms to remove harmful content and let users opt out of algorithm-generated feeds. Platforms could face fines exceeding $100 million, with pressure to increase the penalties further.
The Guardian Technology · 10d ago
A new national standard for AI customer service has taken effect, requiring companies to take responsibility for the systems they deploy. The change aims to address misleading, unhelpful, and unaccountable AI customer-service experiences.
钛媒体 · 12d ago
Representatives from the G20 economies unanimously agreed to adopt US-proposed guidelines favoring a lighter-touch approach to AI regulation. The agreement represents a policy win for the Trump administration and Silicon Valley.
Bloomberg Technology · 17d ago
New York City will impose a one-year moratorium on classroom AI use for students from 2-K through eighth grade during the 2026–2027 school year. The policy also bars teachers from using AI to grade assignments and introduces additional digital-device limits alongside a small AI education pilot.
The Verge · 17d ago
The European Commission has designated ChatGPT, Reddit, and Roblox as very large online services under the Digital Services Act. ChatGPT is classified as a very large online search engine, bringing it under stricter regulatory oversight and scrutiny.
cnBeta (Full RSS) · 19d ago
ChatGPT and Reddit now face the European Union’s toughest online safety requirements. The development reflects how rapid platform growth is bringing increased regulatory obligations in the EU.
Ars Technica AI · 19d ago
OpenAI will terminate Cursor’s direct access to its models on November 12, citing concerns that SpaceX may not comply with its service terms after acquiring Cursor. OpenAI will not provide Cursor with future models, although the contract termination date is being extended to the latest permissible date.
极客公园 · 20d ago
The Trump administration has circulated a draft executive order proposing an industry self-regulatory organization for leading AI model companies. The plan still requires approval from senior officials, including the president, and progress has recently stalled.
cnBeta (Full RSS) · 22d ago
Taiwanese prosecutors charged nine people with illegally exporting high-end B300 AI servers to mainland China. The defendants include one Nvidia employee and two Super Micro employees, highlighting escalating enforcement around AI hardware restrictions.
The Guardian Technology · 25d ago
OpenAI is calling on California to strengthen its AI safety framework under SB 53. The company says the bill should be amended to expand safeguards.
Engadget · 28d ago
OpenAI reaffirmed Zero Data Retention for eligible API customers using frontier models. It also previewed Private Safety Processing, designed to support advanced AI safety work without compromising customer data privacy.
OpenAI News · 31d ago