China's Qwen3 AI chatbot accused of rewriting Tiananmen Square history - Daily Sun
Britain's outgoing terrorism watchdog has warned that China's artificial intelligence ( AI ) technology poses a threat to Western democracy ...
Open-Source AI 3,556
Open-source and open-weight models, and the tools and community around them.
Related topics: China & GeopoliticsCybersecurityChatbots & LLMsRegulation & PolicyMarkets & InvestmentAI Safety & Existential RiskAI Agents & AutomationBanking & Finance
Chinese AI models show alternative path to high performance at lower cost: The Hindu director
Chinese AI models such as DeepSeek and Kimi have offered an alternative approach to AI development, demonstrating that high-performing models ...
Mistral pushes back on new AI copyright proposals, sparking outcry from creatives | Sifted
The Paris-based AI model maker is urging parliamentarians to vote against a new proposal aiming to tackle copyright issues in generative AI .
Copyright & IPOpen-Source AIRegulation & PolicyChatbots & LLMs
Cyberattacks threaten to make open-source AI a little less open, Hugging Face exec warns
The risks aren't limited to platforms that invite open community contributions. On Monday, AI infrastructure startup Fireworks, which helps ...
OPINION | ART HOBSON: AI's dangers create threats | Northwest Arkansas Democrat-Gazette
Recent events suggest a novel addition to this list: runaway artificial intelligence . AI is a real danger , a point made by the Hugging Face incident ...
DeepSeek Restrictions DOJ Whistleblower Rewards GAO Bid Protests and Buy America Waivers
Recent developments in government contracting include new restrictions on the use of certain AI technologies in federal systems, a US Department ...
Governance & Risk ManagementAI Safety & Existential RiskGovernment & Public SectorOpen-Source AI
Chinese AI labs hire Silicon Valley engineers to rival OpenAI, Anthropic - Rest of World
Lead development and training of advanced AI models. MiniMax, AI Data Center Supply Chain Lead, San Francisco, Manage AI computing resources and chip ...
Terror tsar: China's AI bot is rewriting history - The Telegraph
Terror tsar: China's AI bot is rewriting history. Jonathan Hall downloaded Qwen3 to his wife's laptop. It told him the Tiananmen Square massacre never ...
Democrats weigh the future of the House China committee that drove chip export-control scrutiny
... China . In April 2025 it called Chinese AI lab DeepSeek a "profound threat " to U.S. security, said the company may have evaded export controls ...
How AI is breaking cybersecurity | Decoder - BigGo Finance
AI slowdown debate AI midterm politicsOpen-weight model threatsChina AI competition AI ... Threat , Open vs Frontier AI , AI Cyber Warfare & SaaS Collapse.
CybersecurityChina & GeopoliticsBanking & FinanceMilitary & Defense
Who Owns DeepSeek, The Chinese AI Company? - BGR
Chinese AI lab DeepSeek made headlines and caused disruption in the AI industry when it launched the R1 model in 2025. It uses a concept called ...
AI agent incident at OpenAI and Hugging Face: what happened | Consulting | RSM Panamá
Find out what happened in the AI agent incident involving OpenAI and Hugging Face, and what risks it reveals for the adoption of autonomous AI .
State propaganda in some Chinese AI models is only "a thin layer" that fine-tuning can strip away
Nous Research retrained Alibaba's Qwen3.8-27B to produce Arete-27B, largely stripping out its pro-Chinese bias . The propaganda rate dropped from ...
Bias & FairnessMisinformationChina & GeopoliticsOpen-Source AI
With the use of open AI models expanding rapidly, what are the differences between ... - GIGAZINE
'Mistral Large 4,' an AI model that delivers coding performance equivalent to GPT-6.1 Terra, has been released free of charge. Trinity-Large ...
VAPOR PANDA Using AI Pen Testing Platform to Enhance Hacking Speed | JD Supra
... Chinese -developed DeepSeek LLM to attack “more than 200 distinct ... When an AI Chatbot's Reassurance Becomes a Risk · Privacy Tip #515 ...
China & GeopoliticsCybersecurityChatbots & LLMsPrivacy & Surveillance
Free Chinese AI Models Are Getting Scary Good at Hacking | WIRED
An experiment in which AI models were asked to break into simulated networks showed the hacking skills of DeepSeek and GLM to be not far behind ...
China's AI models, led by Kimi, dominate open-source enterprise spending
The report, released this week by Silicon Valley venture capital firm Menlo Ventures, said Chinese AI systems accounted for 89 per cent of global ...
Cisco Open Sources Project Quarry To Detect Dangerous AI Agent Behaviour Across Entire Sessions
Cisco has open sourced Project Quarry, a detection language and engine designed to identify harmful behaviour across AI agent sessions, ...
AI Safety & Existential RiskAI Agents & AutomationOpen-Source AI
Qwen Office E-commerce Edition Field Test: 3-Hour Store Setup Success, But Facing Image ...
... AI hallucinations . For a research report that takes data rigorously as its selling point, wrong brand names and disordered data not only cannot ...
Hallucinations & FabricationOpen-Source AIRetail & E-commerce
# AI : Neither Tortoise Nor Hare! - Yahoo Tech
Out-of-control artificial intelligence ( AI ) agents are making headlines. The widely publicized attack on Hugging Face by OpenAI agents has been ...
Chinese AI Firm DeepSeek Advances Super-Agent Design To Cut Human Oversight In Multi ...
Chinese AI company DeepSeek has outlined a new architectural approach for AI agents that aims to reduce the amount of direct human coordination ...
China & GeopoliticsAI Agents & AutomationOpen-Source AIGovernance & Risk Management
Global Times: China champions open AI co-op at Wuzhen Summit, calls for opposition to ...
... Chinese software and AI solutions could be applied in Sri Lanka. ... China's rapid advances in AI , particularly its expanding open-source ...
Why AI makes your open-source stack riskier than ever - Spiceworks
AI -driven threats make open-source vulnerabilities a critical enterprise risk . Reevaluating supply chains and SBOMs is now essential for security.
Hugging Face CEO just told Sam Altman and Dario Amodei - The Times of India
... AI risks , real safety requires accountability and open development rather than corporate secrecy. He added that the biggest dangers surrounding AI ...
China's Tencent Hy4 challenges world's top models with 7.2% error rate
... hallucination rate from 12.95% to 7.2%, according to internal evaluations. China's Tencent has introduced a new open-source AI model that can ...
Hallucinations & FabricationChina & GeopoliticsMarkets & InvestmentOpen-Source AI
Iran and the AI Threats We Cannot See | The Washington Institute
A growing body of evidence shows how Iran and its proxies are using artificial intelligence and especially more easily manipulatable open-weight ...
Tencent Launches Hy4, Joining the Ranks of the World's Leading Open-Source AI Models
... Chinese market. In practical use, Hy4 balances efficiency and effectiveness for visual creative tasks, 3D scene construction, and web development ...
Tencent Declares Latest AI Model Ranks Among China's Best - Bloomberg.com
Tencent Holdings Ltd. released a foundation model Wednesday that it said joins the top tier of Chinese open-weight AI systems for the first time.
Hugging Face's CEO Says His P(Doom) Is 'Close to Zero' - WIRED
Most of the risk is actually created by a couple of players building behind closed doors,” Clément Delangue said of the need for open source AI .
DeepSeek engineer defends push for open-source AGI despite higher risks
Some AI pioneers have warned that technology smarter than humans may escape human control.