Amazon Overhauls AI Strategy as Flagship Nova Models Wind Down
finance.yahoo.comAmazon redirects resources toward a new frontier model after struggling to generate interest around several Nova products, marking a strategic pivot in their AI portfolio.
Gemini is moving into Google Play, and it has front-row seat to your spending habits
digitaltrends.comGemini is being integrated into Google Play and Google Pay to analyze spending patterns, explain financial topics, and provide a front-row seat to users' financial information. A new partnership with SBI expands these capabilities further.
Gemini on Mac can now type what you say and act on what it sees
digitaltrends.comGoogle is upgrading the Gemini app for Mac with new features including system-wide dictation triggered by Fn key and an opt-in mode to pull context from on-screen content. These enhancements expand what users can do through their AI assistant.
Tesla's Grok Playbook Meets Its Match? GOOGL's Waymo Arms Ojai Robotaxis With Gemini AI
finance.yahoo.comGoogle is integrating its Gemini model into Waymo's robotaxi fleet, where it provides voice and touch controls for cabin settings in autonomous vehicles. This demonstrates real-world deployment of the AI model beyond Google ecosystem products.
Claude's 'Private' Share Links Pop Up In Google Search
ijr.comUsers who share Claude chats with the option to create public links may find their conversations appearing in Google search results.
Research: How AI Agents Broaden the Scope of Knowledge Work - Perplexity field study
hbr.orgA field study examining how AI agents at Perplexity change the nature of knowledge work and expand its scope, with insights into their practical application.
Elon Musk's xAI sues Minnesota over its first-in-the-nation law banning nudification technology
boston.comxAI sues Minnesota over legislation prohibiting nudifying AI tools, which the company may view as a critical feature of its Grok model.
OpenAI launches free AI access program for academic researchers
tech.yahoo.comOpenAI is launching a new program to provide 100,000 academic researchers with free access and featuring improved safety filters while maintaining open weights transparency.
OpenAI faces mounting lawsuits, state scrutiny over ChatGPT Health - Becker's...
beckershospitalreview.comOpenAI is facing product liability lawsuits and state-level data probe scrutiny related to ChatGPT Health, which has encountered AI safety concerns in healthcare applications.
Samaya AI Releases FrontierFinance, a New Public Benchmark for AI Agents in...
fox44news.comSamaya AI launched FrontierFinance, a public benchmark evaluating AI agents in investment workflows outperforming models including Claude Fable 5 and GPT-5.6 SOL.
Evinced Harnesses Agentic Coding to Automatically and Reliably Fix Accessibility...
prnewswire.comEvinced leverages agentic coding AI tools to automatically identify and remediate accessibility issues in web applications.
Cursor cut its India price to $7, and its own AI models are why
thenextweb.comCursor launched a new ₹649/month (approx. $8) India-only plan built on its own AI models instead of OpenAI or Anthropic's, charging about one-third the price of Pro tier while including premium features like Composer and cloud agents.
Here's How Microsoft's Mistral Partnership Could Strengthen Azure's AI Leadership
finance.yahoo.comMicrosoft and Mistral expanded their multibillion-dollar partnership to strengthen Azure's AI leadership position for enterprise customers.
Gemma 4's quantized models finally made local AI practical in my homelab
msn.comAn article about Gemma 4's quantized models enabling practical local AI deployment in home labs, demonstrating their speed and accuracy improvements.
Liquid AI、CPUで長文を高速解析する「LFM2.5-Encoders」公開
msn.comLiquid AI 公开发布了可在CPU上快速处理长文的AI模型「LFM2.5-Encoders」，提供两种尺寸（230M/350M）的开源权重供用户根据需求选择。
I gave Claude Opus 5 and Kimi K3 15 impossible prompts — the winner surprised me
msn.comA comparative test pitting Claude Opus 5 against Kimi K3 with demanding coding and infrastructure prompts to see which AI model performs better.
How a Chinese AI model stopped OpenAI's 'unprecedented' cyber attack
msn.comCNBC's Kai Nicol-Schwarz discusses how Chinese AI model GLM 5.2 was used by Hugging Face to defend against a cyber attack, highlighting the role of open-source models in security scenarios.
Nvidia, Meta: China's AI Shouldn't Be Banned—Here's Why - Meta Platforms (NASDAQ...
benzinga.comNvidia and Meta argue that China's AI models shouldn't be banned, stating open-source AI strengthens global innovation. The article discusses the companies' stance on Chinese LLM competition in July 2026, referencing Meta's strategy around open-model collaboration including llama ecosystem.
How to make money online with Google Veo 3: 5 proven AI video strategies
msn.comGuide on monetizing Google's new AI video generator Veo 3 with five proven strategies for using the model in business.
The 4 qualities that make the Moonshot founder so exceptional, according to his Ph.D advisor
msn.comMoonshot AI founder recognized for qualities shaping his approach to how large-scale AI models should work, based on feedback from former Ph.D advisor.