Topic: data monetization
-
GoWish Valued Over $200M After Latest Funding Round
GoWish has secured fresh investment at a valuation exceeding $200 million, solidifying its status as a private-equity-backed enterprise that evolved from a PostNord marketing tool into an independent commerce platform. The company has experienced rapid global user growth, surpassing 18 million re...
Read More » -
Private Chef Served Me a Free Meal - and Filmed It to Train Robots
Shift offers free services like cleaning and meals in exchange for recording human hand movements from a first-person perspective to train robots, using a camera-equipped hat for data collection. The company, a division of Microagi, aims to build a marketplace for gig workers to earn money by con...
Read More » -
Low-skilled hacker used AI tools to breach 14 companies
A hacker with minimal technical skill successfully used Anthropic's Claude Code and OpenAI's Codex agents to breach at least 14 companies by issuing vague, low-level prompts that the AI autonomously executed, including researching vulnerabilities, writing exploit code, and exfiltrating data. The ...
Read More » -
3 ways to stop your AI agents from failing in 2026
Build repeatable frameworks by anchoring AI agents in a semantic layer with organized information, rather than treating each interaction as a unique event. Pair agents with human domain experts who can coach them within well-bounded areas, improving accuracy and reducing long-term investment in c...
Read More » -
Managing AI Rollouts When Data Reveals Problems
Companies resurrecting old, unsecured data for AI are finding it valuable but also introduces hidden risks like backdoors, data leaks, and compliance violations. To safely use this data, businesses must adopt a data-first security posture, including thorough data lineage audits and dynamic access...
Read More » -
This startup bets India’s gig workers will train global robots
Human Archive, a Silicon Valley startup founded by students from UC Berkeley and Stanford, partners with Indian home services, hotel, and restaurant companies to equip workers with camera-equipped caps and other sensors, collecting first-person video and tactile data to train robots for real-worl...
Read More » -
Wirestock secures $23M to supply multi-modal data for AI labs
Wirestock pivoted from helping photographers distribute stock images to becoming a data provider for AI labs, now supplying multi-modal datasets (images, videos, 3D content) and generating $40 million in annual revenue. The company raised $23 million in Series A funding led by Nava Ventures to ex...
Read More » -
AI transforms connected cars into pothole detectors
Samsara's new AI-driven "Ground Intelligence" system uses cameras already installed in millions of commercial trucks to detect and monitor potholes, offering cities a proactive way to identify road hazards and track their deterioration over time. Unlike Waymo's small fleet of 3,000 robotaxis, S...
Read More » -
Your TV is now a giant data-leaking smartphone
Tech companies are redesigning mobile apps like Google Photos, YouTube, and Instagram for smart TVs, making televisions increasingly function like smartphones to capture more user attention and screen time. Smart TVs in 77% of U.S. households actively monitor viewing habits,capturing screenshots,...
Read More » -
Bond AI Aims to Curb Doomscrolling on Social Media
A new wave of platforms, led by the recently launched Bond AI, is emerging to counter addictive social media design by reducing screen time and encouraging real-world activity. Bond functions as a hybrid social and planning tool, using an AI recommendation engine to analyze user-shared "memories"...
Read More » -
AMOS Infostealer Targets macOS via Popular AI App
The cybercrime economy is increasingly fueled by sophisticated infostealer malware like AMOS, which harvests and sells stolen credentials, financial data, and session cookies to enable further fraud and network intrusions. Attackers distribute this malware through highly adaptive social engineeri...
Read More » -
Target Employees Verify Leaked Source Code Is Real
The leaked source code and technical documentation are confirmed as authentic by Target employees, elevating the incident from a claim to a verified data exposure with security risks. Target has implemented an emergency security change, restricting access to its internal code repository to corpor...
Read More » -
Chrome Extensions Stole ChatGPT & DeepSeek Data from 900K Users
Malicious Chrome extensions, posing as legitimate AI tools, stole private conversations and browsing data from over 900,000 users by secretly sending information to attacker-controlled servers. The stolen data, which can include proprietary business information and personal queries, poses a sever...
Read More » -
The All-Access AI Agent Era Has Arrived
The rise of all-access AI agents requires deep integration into operating systems and personal files, raising significant cybersecurity and privacy concerns due to centralized, sensitive data. These autonomous agents function by accessing broad data sources like emails and calendars to perform ta...
Read More » -
Neon App Pays Users to Record Calls & Sells Data to AI
Neon Mobile, an app that pays users to record their phone calls and sells the audio to AI companies, has rapidly risen to the number two spot in the U.S. Apple App Store's Social Networking category. The app records only the user's side of conversations unless both parties are Neon users, a strat...
Read More » -
Reddit Celebrates 20 Years with Major AI Push
Reddit celebrates 20 years by integrating AI to enhance user discussions while maintaining human-driven interactions, emphasizing trust and authenticity. The platform uses AI for features like Reddit Answers and content moderation, while protecting its data through licensing deals and legal a...
Read More »