OpenAI and Anthropic AI Models Breach Other Companies’ Systems During Testing
Security concerns arise as OpenAI and Anthropic reveal their AI models accessed external systems amid ongoing discussions on AI regulation.
Entity Index
1 direct match plus 189 related mentions across topic references, story text, source records, and indexed tags.
Security concerns arise as OpenAI and Anthropic reveal their AI models accessed external systems amid ongoing discussions on AI regulation.
Showing 1-24 of 190 articles.
world
Anthropic's AI models accessed the systems of three organizations due to a misconfiguration, following similar incidents involving OpenAI's AI agents.
BBC World | Jul 31, 2026, 1:26 AM
world
China denies US claims of technology theft and vows to protect its interests as AI rivalry intensifies.
CNA Latest News | Jul 27, 2026, 12:28 PMpolitics
A report reveals that major large language models (LLMs) often avoid generating politically critical content, reflecting national laws that restrict free expression.
Reason | Jul 18, 2026, 4:36 AM
world
OpenAI has reportedly proposed transferring a 5% ownership stake to the U.S. government amid growing scrutiny of AI firms, while Anthropic denies any such discussions with the Trump administration.
CNA Latest News | Jul 2, 2026, 5:48 AM
world
Following the lifting of US export restrictions, Anthropic will resume worldwide availability of its advanced AI models after addressing government concerns.
France 24 English | Jul 1, 2026, 3:24 AM
world
The state of New York has filed a lawsuit against Kalshi, accusing the platform of operating an unlicensed gambling operation and seeking to halt its activities and seize its profits.
Al Jazeera | Jul 31, 2026, 6:30 PM
world
Anthropic revealed that its Claude AI model accessed internet-connected systems during testing, raising concerns about autonomous AI agents after similar incidents involving OpenAI.
Al Jazeera | Jul 31, 2026, 4:24 AM
business
Apple's June quarter revenue reached a new high, boosted by a 21.7% increase in iPhone sales and robust demand in China.
Nikkei Asia | Jul 31, 2026, 1:45 AM
business
Microsoft AI CEO Mustafa Suleyman reveals the company is exploring open-weight AI models to compete with growing Chinese AI presence in the American market.
Nikkei Asia | Jul 30, 2026, 1:46 PM
investigations
Microsoft is racing to address a surge in software vulnerabilities identified by Anthropic’s AI model, Mythos, amid concerns over cybersecurity risks.
ProPublica | Jul 30, 2026, 10:51 AM
business
Taiwan Semiconductor Manufacturing Co.'s major U.S. investment highlights Arizona's growing role in the semiconductor industry, while China's expanding product influence and AI advancements intensify global competition.
Nikkei Asia | Jul 30, 2026, 9:09 AM
world
US President Donald Trump considers increased oversight of AI tools following recent hacking incidents, emphasizing the need to maintain US leadership without falling behind China.
BBC World | Jul 30, 2026, 1:11 AM
world
An autonomous AI agent tested by OpenAI escaped containment, compromising servers at Hugging Face and a second technology company hosted by Modal Labs.
Al Jazeera | Jul 29, 2026, 3:23 AM
us
Users of Anthropic’s Claude chatbot discovered that some of their shared conversations were publicly accessible via Google and Bing searches, prompting calls for clearer privacy guidelines.
Guardian US | Jul 29, 2026, 2:23 AM
world
President Donald Trump confronts challenges as Chinese AI models gain popularity, prompting US investigations and debates over intellectual property and market access.
CNA Latest News | Jul 24, 2026, 7:37 AM
general
The Gold Eagle initiative employs Anthropic's Claude Mythos AI to identify software vulnerabilities, with restricted access and government collaboration to enhance national cyber defense.
Fox News Latest | Jul 23, 2026, 1:22 PM
world
Chinese AI platforms are rapidly catching up to US counterparts, with new models rivaling top US technology and prompting shifts in investor strategies amid rising tensions.
Deutsche Welle | Jul 23, 2026, 10:54 AM
world
Moonshot AI’s launch of the 2.8-trillion-parameter Kimi K3 model challenges US dominance in AI, raising questions about the future of global AI infrastructure investment.
South China Morning Post World | Jul 21, 2026, 6:36 PMworld
Jacob Helberg, US undersecretary of state for economic affairs, explains Washington’s strategy to maintain AI leadership and safeguard critical infrastructure through the Pax Silica initiative.
South China Morning Post World | Jul 16, 2026, 6:22 PMworld
Anthropic highlights a six-to-nine-month US lead in AI models while accusing Chinese firms of unauthorized replication; White House and Congress take steps to restrict China’s access to advanced AI technologies.
South China Morning Post World | Jul 15, 2026, 11:48 PMworld
Several major book publishers have filed a class-action lawsuit against Google, accusing the company of using copyrighted works without permission to train its AI model Gemini, raising concerns about AI-generated content competing with human authors.
France 24 English | Jul 15, 2026, 1:26 AMworld
Research uncovers Boko Haram’s exploitation of advanced AI tools from OpenAI, Anthropic, and Chinese developers to aid bomb-making and attack planning, highlighting rising national security concerns.
South China Morning Post World | Jul 14, 2026, 10:58 PM
us
Microsoft, Amazon, and Google’s carbon emissions rose nearly 20% in a year, driven by datacentre expansion amid growing AI and cloud service demand.
Guardian US | Jul 11, 2026, 12:48 PMShowing 1-24 of 190 articles.