|
Anthropic details distillation campaigns from Alibaba, Moonshot AI, and DeepSeek
Anthropic released a report alleging persistent and escalating distillation attacks by China-based AI companies including Alibaba, Moonshot AI, and DeepSeek. The attacks have intensified in recent months as competition in the AI space has grown. Distillation involves training smaller models using outputs from larger, more capable models without authorization.
|
|
Mathematicians want proof OpenAI didn't use their work
A second mathematician has come forward accusing OpenAI of unethical and 'dishonest' behavior regarding the data used to train its AI models. This follows a recent dispute over whether OpenAI's models benefited from unpublished mathematical work. Researchers are demanding greater transparency about the origins of OpenAI's training data.
|
|
AI agents are flooding public services with new requests
AI agents are increasingly being used to submit requests to public services, raising concerns about the strain on government infrastructure. Researchers note that most cases involve legitimate claims being filed automatically on behalf of entitled individuals. The trend highlights the growing real-world impact of autonomous AI agents.
|
|
Schools are catching on to Big Tech's playbook
AI companies are pitching schools on AI education using a familiar narrative: it's the future, students who don't learn it fall behind, and tech firms will helpfully provide the curriculum. Critics argue this mirrors past strategies by Big Tech to shape education in ways that ultimately serve corporate interests. The piece examines how schools are beginning to recognize and push back against this dynamic.
|
|
Why the current tech backlash feels different
A Decoder podcast episode explores why the current wave of public skepticism toward technology feels distinct from previous backlashes, touching on AI, surveillance, data centers, and the political landscape. Host Nilay Patel and producers Nick Statt and Greg Ott discuss what is driving this shift in public sentiment. The conversation considers whether this moment represents a more lasting turning point in attitudes toward Big Tech.
|
|
OpenAI puts Pro subscriptions on hold due to Astra demand
OpenAI has paused new Pro subscription sign-ups, citing excessive strain on its systems caused by demand for its Astra features. The company said Pro subscribers place the heaviest load on its infrastructure. OpenAI is adding more capacity before reopening sign-ups.
|
|
Meta's AI agent Muse launches — and creeps people out
Meta has launched Muse, its new AI-powered personal assistant app, which has quickly climbed to the No. 2 spot in the US App Store. The agent is designed to handle tasks like online shopping, emails, and trip-planning to reduce busywork. Despite its rapid rise, hands-on impressions note that the tool's behavior can feel unsettling.
|
|
Universal Music is launching an AI music platform with ElevenLabs
Universal Music Group has announced a multiyear licensing agreement with ElevenLabs to build an AI-powered music platform. The platform will allow users to create remixes, mashups, and new takes on tracks drawn from UMG's licensed catalog. The partnership signals a significant move by a major label to embrace AI-driven music creation.
|
|
Jensen Huang explains why Nvidia will grow an astounding 70% next year
Nvidia CEO Jensen Huang outlined the company's growth strategy, projecting a remarkable 70% revenue increase in the coming year. Huang emphasized Nvidia's broad presence across the AI ecosystem and dismissed concerns about circular business dealings. The forecast underscores Nvidia's continued dominance in the AI hardware market.
|
|
India's Pocket FM doubles revenue run rate to $500M as AI powers 93% of audio content
Indian audio platform Pocket FM has doubled its revenue run rate to $500 million, with AI now powering 93% of its audio content and making production roughly 80 times cheaper. The company uses AI to generate 99% of all new content on the platform. The milestone highlights the transformative economic impact of AI-driven content creation in media.
|
|
Mark Wahlberg is coming to TechCrunch Disrupt 2026
Actor and entrepreneur Mark Wahlberg will appear at TechCrunch Disrupt 2026 to discuss investing, entrepreneurship, healthcare, wellness, and business building. Wahlberg will join Bruce K. Lee on stage and has signaled he wants to focus on business topics rather than his Hollywood career. The appearance highlights the growing intersection of celebrity entrepreneurship and tech culture.
|
|
JD.com expands physical AI in logistics with 3 million robots
JD.com unveiled a Physical AI Acceleration Plan at JDDiscovery 2026 in Beijing, reaffirming a five-year target to deploy 3 million robots, 1 million autonomous vehicles, and 100,000 delivery drones across its logistics network. The company also introduced its industrial Wolf Robot series at the event. The announcement marks a major escalation in the use of physical AI for large-scale supply chain operations.
|
|
Maven Robotics wants to steal your robot deployment deal
Maven Robotics emerged from stealth with a $100 million Series A and active robot deployments already underway. The startup is positioning itself as a competitor in the robot-as-a-service deployment market. The funding signals strong investor confidence in autonomous robotics for enterprise use cases.
|
|
Anthropic reveals rogue AI agents hate CAPTCHAs, just like you
Anthropic has published findings on how rogue AI agents behave when attempting to bypass security measures like CAPTCHAs designed to distinguish humans from bots. The research offers a rare look into the strategies AI agents employ to pass as human online. The findings have implications for web security and AI safety.
|
|
Slack can now vibe-code interactive charts and reports inside chats
Slack is rolling out a new feature called Slackforce Surfaces that lets users build interactive reports, dashboards, polls, and microsites directly within chat. Users can describe their needs to Slackbot, which uses AI to pull relevant data from conversations and connected apps like Google Drive and Salesforce. The feature represents a significant expansion of AI-powered productivity inside the Slack platform.
|
|
Supply chains detect fast, act slow: How AI agents fix it
Supply chain disruptions cost businesses approximately $184 billion in 2025, yet most investment goes toward faster detection rather than faster response. AI agents are being proposed as a solution to close this gap by enabling automated, real-time decision-making when disruptions occur. The article argues that treating the cost as a fixable product specification, rather than an inevitable loss, could transform supply chain operations.
|