|
OpenAI's Hugging Face breach reignites debate over AI alignment and control
A breach involving OpenAI and Hugging Face has reignited debate over AI alignment and control, exposing competing views on whether increasingly capable AI should be better aligned, better contained, or both. The incident highlights growing tensions in the AI safety community over how to manage frontier model risks.
|
|
Nvidia and Microsoft launch Open Secure AI Alliance for cybersecurity — without OpenAI, Google, or Anthropic
Nvidia has joined forces with Microsoft, SpaceX, IBM, and other tech companies to form the Open Secure AI Alliance, aimed at building and sharing open-source AI security tools. The initiative is a direct response to mounting concerns over safety threats posed by frontier AI models. Notably, major AI labs OpenAI, Google, and Anthropic are absent from the alliance.
|
|
America's AI investment boom is reshaping the economy
Microsoft, Meta, Amazon, and Alphabet are collectively committing hundreds of billions of dollars to AI infrastructure, making artificial intelligence one of the defining investment stories in the US. Surging demand for advanced chips has turned Nvidia into one of the most valuable companies in the world. The scale of investment is beginning to reshape broad economic dynamics across sectors.
|
|
Satya Nadella says companies that rely on a single AI may not survive
Microsoft CEO Satya Nadella warned that companies without their own AI models or AI gateway infrastructure to separate their prompts from underlying models risk being left behind. He emphasized the importance of maintaining a layer of control over AI systems rather than depending entirely on a third-party provider. The comments underscore a broader strategic shift toward AI sovereignty for enterprises.
|
|
PSA: Your Claude shared chats and Artifacts may have ended up on Google
Anthropic's Claude has a privacy issue: shared chats and Artifacts created via the platform's 'share chat' feature appear to have been indexed by Google, making them publicly discoverable. The feature allows users to generate links that let anyone view a conversation or project, but many users may not have realized these links were being crawled. Users are advised to review and delete any shared chats they consider sensitive.
|
|
Why China is giving away its best AI models
Silicon Valley is on high alert following the arrival of Moonshot AI's Kimi K3, a Chinese open-weight AI model that allegedly rivals top US systems at a fraction of the cost. The release intensifies the US-China AI rivalry and raises strategic questions about why Chinese companies are releasing powerful models openly. Anthropic CEO Dario Amodei has also weighed in, saying he does not oppose open-weight models in general but specifically fears the rise of Chinese AI capabilities.
|
|
Google's AI search is rapidly becoming the default, new data shows
Google's AI Overviews now appear in 43% of searches, according to new data, underscoring how quickly AI-generated answers are becoming the default way people discover information online. The rapid adoption raises questions about the future of traditional web traffic and publisher revenue. It marks a significant shift in how the world's dominant search engine delivers results.
|
|
Meta AI is now available in Threads DMs
Meta has begun rolling out its Meta AI chatbot directly within Threads' direct messages, giving users a new way to interact with the AI assistant inside the social platform. The integration expands Meta AI's footprint across the company's family of apps. It follows similar AI assistant integrations across WhatsApp and Instagram.
|
|
Microsoft launches its first cybersecurity AI model and a new agentic security platform
Microsoft has bolstered its AI security offerings with the launch of its first dedicated AI security model and a new agentic cybersecurity system. The announcements position Microsoft as a major player in AI-driven cyber defense. This move aligns with broader industry concerns over AI-enabled threats.
|
|
TechCrunch Disrupt 2026: Smart Systems Stage to focus on AI infrastructure and energy
TechCrunch has released the agenda for the Smart Systems Stage at Disrupt 2026, which will cover AI infrastructure, energy challenges, and topics ranging from fusion breakthroughs to the grid strain caused by AI's growing compute demands. The stage aims to bring together leaders at the intersection of energy and technology. It reflects growing industry awareness of AI's physical infrastructure requirements.
|
|
This $9 NFC key physically locks your most addictive apps
A new $9 NFC key device requires users to physically scan it in order to unlock distracting or addictive apps on their phone, offering a tangible barrier to mindless scrolling. The low-cost gadget is aimed at people looking for stronger screen time controls than software-based solutions provide. It represents a novel approach to digital wellness through physical friction.
|
|
Ilya Sutskever's Safe Superintelligence partners with Nvidia to scale AI research
After two years operating in stealth, Safe Superintelligence has announced a long-term partnership with Nvidia as it moves into its next phase of scaling AI research. The partnership signals that SSI is preparing to emerge from stealth with significant compute backing. Founded by former OpenAI chief scientist Ilya Sutskever, the company has been one of the most closely watched AI startups.
|
|
How AI is shortening drug discovery timelines in China
Insilico Medicine has reduced the time needed to produce drug development candidates to about one year by combining AI with laboratory research, with its fastest programme reaching candidate nomination in just nine months. CEO Alex Zhavoronkov says the approach is transforming what was previously a multi-year process. The Hong Kong-listed company is positioning itself at the forefront of AI-driven pharmaceutical development.
|
|
Enigma raises $71M to make robot control as intuitive as adjusting the volume
Robotics startup Enigma has raised a $71 million seed round led by Index Ventures and Ribbit Capital, with participation from Sarah Guo's Conviction Partners, to develop more intuitive robot control interfaces. The company aims to make operating robots as simple and natural as adjusting a device's volume. The massive seed round reflects continued investor appetite for robotics startups leveraging AI.
|