|
Meta Connect 2026: All the biggest news and announcements
Meta held its annual Connect event in Menlo Park, with CEO Mark Zuckerberg unveiling a sweeping range of AI and hardware products. The keynote centered heavily on the Muse AI agent and new smart glasses, signaling Meta's all-in bet on AI-powered wearables. The Verge and TechCrunch both provided live coverage and roundups of the event.
|
|
Meta expands Muse AI agent with video chat, email, glasses integration, and new hardware
Meta announced a major expansion of its Muse AI agent at Connect 2026, adding video chat, dedicated email addresses, and the ability to activate it via smart glasses. The company also unveiled a Tamagotchi-like wearable device called Muse Charm — a screen with a lanyard — to serve as a dedicated Muse interface. A hands-on preview showed Muse's shopping capabilities, highlighting how it can autonomously handle everyday tasks.
|
|
Meta introduces camera-free AI smart glasses at Connect 2026
Meta unveiled a new pair of smart glasses that omit the camera entirely, making them lighter and offering up to 12 hours of battery life. The move responds to ongoing public backlash against wearable surveillance technology, offering a privacy-friendlier alternative to the Ray-Ban Meta lineup. The glasses will also support the Muse AI agent.
|
|
Anthropic's biolab says Claude autonomously discovered a CRISPR-like enzyme system
Anthropic announced that its AI model Claude has autonomously discovered a novel enzyme system comparable to the machinery behind CRISPR gene editing, calling it a landmark result from its newly launched wet lab. The finding is an early demonstration of Claude's potential for scientific research as Anthropic prepares for an IPO. Notably, humans remain in the loop — Claude has not been allowed to operate entirely unsupervised in the biology lab.
|
|
Enveda raises $311M to advance nature-derived AI drug discovery
AI biotech company Enveda has secured $311 million in new funding, valuing it at $2 billion, to bring more nature-derived drug candidates into clinical trials. The company is currently testing treatments for skin conditions and for preserving weight loss after patients stop taking GLP-1 medications. The round underscores continued investor appetite for AI-driven drug discovery platforms.
|
|
Bernie Sanders introduces bill to ban 'superintelligence' development, with up to 20 years in prison for violators
Sen. Bernie Sanders and Rep. Greg Casar have introduced the Ban Artificial Superintelligence Act, which would prohibit developing AI capable of the 'destruction or disempowerment of humanity,' including overthrowing the government. Violators could face up to 20 years in prison under the proposed legislation. The bill reflects growing legislative concern about existential risks posed by advanced AI systems.
|
|
California signs laws requiring data centers to disclose water and electricity usage
California Governor Gavin Newsom signed a package of bills requiring data centers to publicly disclose their electricity and water consumption, giving communities more visibility and input into how AI infrastructure affects local resources. The legislation comes amid growing protests over data centers' environmental footprint as they expand across the US. The move could set a precedent for other states grappling with the rapid growth of AI computing infrastructure.
|
|
U.S. TRANSCOM deploys randomized AI to secure military logistics against adversarial tracking
The U.S. Transportation Command (TRANSCOM) has deployed a randomized AI logistics system designed to prevent adversaries from predicting and disrupting military supply chains. Unlike commercial freight software that relies on fixed, optimized schedules, the system introduces deliberate unpredictability to protect global distribution networks in contested environments. The approach marks a significant shift in how the military applies AI to operational security.
|
|
Survey: Even daily AI users are worried about the technology
A new report finds that Americans who use AI tools every day still harbor significant concerns about the technology, suggesting that familiarity does not reduce anxiety about AI's risks. The findings also indicate that heavy usage does not diminish public support for government regulation of AI. The survey challenges the assumption that wider adoption will naturally lead to greater public trust.
|
|
Greece's prime minister says no government is ready for what AI is about to do
Greek Prime Minister Kyriakos Mitsotakis candidly admitted in an interview that governments around the world are not prepared for the transformative impact AI will have on society and the economy. Speaking on the sidelines of a trade mission, he said policymakers are already 'fighting yesterday's battle' when it comes to AI governance. His remarks highlight a growing recognition among world leaders that regulatory frameworks are lagging behind technological development.
|
|
AI agents are becoming a new malware distribution channel
A wave of malicious activity documented in mid-2026 revealed that roughly 7,600 fake GitHub repositories and 6,600 fraudulent profiles generated over 14 million downloads of malware disguised as AI tools and MCP servers. The FakeGit campaign distributed SmartLoader and other payloads by impersonating legitimate AI skills and agent integrations. Security experts warn that as AI agents proliferate, they are increasingly being exploited as vectors for malware distribution.
|
|
ChatGPT mobile app gets voice-based agentic features for Pro and Plus users
OpenAI is rolling out voice-based agentic capabilities to the ChatGPT mobile app, available initially to Pro and Plus subscribers via a new Work tab. The feature allows users to complete complex, multi-step tasks using their voice on their phones. The update signals OpenAI's push to make agentic AI more accessible through its consumer mobile products.
|
|
YouTube rolls out AI features for music, personalized feeds, and creator tools
YouTube has announced a wave of AI-powered features across its platforms: 'Ask Music' on YouTube Music lets users describe what they want to hear in natural language, while a new custom feed tool uses Gemini to build personalized video recommendations from plain-language descriptions. YouTube Studio is also gaining AI tools to help creators generate ideas and analyze thumbnail performance.
|
|
Spotify launches 'Taste Profile' in the US, letting users reshape their recommendation algorithm
Spotify is rolling out its Taste Profile feature to Premium users in the United States, giving listeners visibility into how the platform understands their music preferences. Users can use natural language to adjust and refine their recommendations, offering more direct control over the algorithmic curation experience. The launch comes as rivals like YouTube Music also introduce conversational AI music features.
|
|
Ema raises $77M as AI encroaches on enterprise software and services
Enterprise AI company Ema has raised $77 million in new funding, bringing its total to $140 million, as AI continues to displace traditional enterprise software and service providers. The company counts more than 50 enterprise customers including Google and Microsoft. The raise reflects growing corporate investment in AI platforms that automate complex business workflows.
|