Tech·6,158 stories
Tech News — Page 87
Princeton Study: AI Agents Lack Creativity for Self-Improvement
A Princeton-led study found that AI agents using Anthropic's Claude Opus 4.8 could handle the engineering of AI research but failed to produce publishable papers on open-ended questions, suggesting re
Etched Raises $700M at $21B Valuation Led by Jane Street
AI inference chip startup Etched raised $700M at a $21B valuation in a round led by Jane Street, its new server rack customer, more than doubling its $10.3B valuation from a $300M round in July.
U.S. Army Research Built China's Unitree Robot Dogs
China's Unitree Robotics built its dominant, low-cost robot dogs using breakthroughs from openly published U.S. Army-funded research — a case study in how America's open-innovation model is being outm
We still don’t know how people are really using AI
An independent research project called the AI Observatory analyzed 85,000+ conversations across 52 AI models and found that official usage reports from companies like Anthropic and OpenAI significantl
Women hold quarter of new U.S. AI hires
Women held just over a quarter of new U.S. AI hires last year, compared with roughly half of new hires in non-AI roles, according to new LinkedIn data.
Rezolv raises $12.5M led by Norwest for lending AI
Mumbai-based lend-tech startup Rezolv has raised $12.5 million in a funding round led by Norwest Venture Partners to deepen its use of artificial intelligence across the lending lifecycle.
3M Expert Witness Used ChatGPT in Houston Explosion Lawsuit
An expert witness hired by 3M in a wrongful-death lawsuit over a 2020 Houston explosion used ChatGPT to draft significant portions of his expert report, and his revealing AI prompts — instructing the
Fast.ai Co-founder Returns to AI at Answer.AI
The co-founder of fast.ai is returning to AI at Answer.AI after years away to study microbiology, joining a team building SolveIt — a tool designed around a 1945 mathematician's problem-solving playbo
Relay Shuts Down, CEO Jacob Bank Joins Google Chrome as VP
Workflow automation startup Relay is shutting down after four years, with founder and CEO Jacob Bank returning to Google as VP of Product for the Chrome browser.
Nvidia to Back Ohio Data Center With as Much as $105 Billion
Nvidia said Monday it will commit up to $105 billion to back a massive Ohio data center that OpenAI will lease for 20 years, in one of the chipmaker's largest-ever financings for an AI lab.
GPT-5.6 Sol Pricing Cut by 50%
GPT-5.6 Sol, OpenAI's flagship GPT-5.6 series model, appears on OpenRouter at $2.50 per 1M input tokens and $15 per 1M output, with a 1M context window and multi-provider routing.
AI Slop Is Everywhere. Spotify, LinkedIn and Others Have Had Enough.
Tech companies including Spotify, LinkedIn, and YouTube are scrubbing millions of AI-generated spam tracks, channels, and posts from their platforms, deploying AI detection tools to fight the flood of
Save up to $300 on your TechCrunch Disrupt 2026 pass until August 21
TechCrunch Disrupt 2026 returns to San Francisco's Moscone West October 13-15 with three new AI-themed stages, while discount passes saving up to $300 expire Friday, August 21 at 11:59 p.m. PT.
Flock’s new safeguards leave surveillance loopholes
Flock has updated its license plate reader platform to prevent misuse, but critics argue the changes have loopholes and fail to address broader concerns about mass surveillance and data retention.
Reddit Tests Audio, Video Versions of Posts
Reddit has begun testing a new audio and video format that brings text posts to life, letting users listen or watch narrated versions of popular content in select communities.
Sonic Fire Tech Lands $15M for Sound-Powered Fire System
Sonic Fire Tech raised $15 million to advance its sound-based fire suppression system, which uses infrasonic waves to snuff flames without water or chemicals and the cleanup costs they leave behind.
Nevada Approves Just 10 Tesla Robotaxis
Nevada approved Tesla’s request to operate driverless vehicles in Las Vegas but limited the fleet to just 10 cars, far below the 5,000 it sought, and imposed strict operational constraints.
Anthropic's Safety Frame Becomes Control Moat
An op-ed argues Anthropic wraps its business model in safety language to justify opaque model behavior, anti-competitive access rules, and regulatory capture. The Fable incident — where Claude silentl
NIST AI RMF Emerges as Allied Military AI Trust Yardstick
Allied militaries are rushing AI onto the battlefield, but the interoperability challenge is shifting from access to mutual trust — and a little-known NIST framework is emerging as the only credible c
China Pushes to Become World's Top AI Training Data Supplier
Beijing is pushing to become the world's leading supplier of the text, images, and video used to train AI systems, aiming to counter what it sees as English-language Western bias baked into chatbots l
Get the Tech newsletter
Curated tech stories, every morning. Free.
No spam. Unsubscribe anytime.