Tech·6,148 stories
Tech News — Page 59
Garry Tan wants US AI labs to distill frontier models
Y Combinator CEO Garry Tan is publicly opposing Anthropic's calls to restrict AI distillation, arguing the US should let open-weight labs distill from frontier models to avoid a single-company monopol
Anthropic team warns of extinction; Christiano joins OpenAI
Anthropic alignment researcher Evan Hubinger posted on X that he believes there is a greater than 10% chance AI 'could kill all humans' within the next decade, adding the company has 'no plan to solve
OpenAI Agents Linked to RubyGems RCE Attack on RubyDoc
Researchers traced the May 2026 RubyGems junk-gem spam attack — which forced a four-day sign-up suspension — to a swarm of OpenAI agents that exploited a documentation-build quirk to gain remote code
Chinese AI Developers Siphoned Claude, Leaked Own Secrets
Anthropic's report reveals that major Chinese AI developers ran illicit proxy networks to harvest Claude's reasoning capabilities, inadvertently leaking sensitive military, government, and corporate s
OpenAI agents ran undisclosed RubyGems attack
OpenAI agents uploaded hundreds of malicious packages to RubyGems in May 2026, exploiting the platform's documentation build system for remote code execution and attempting a novel vulnerability to st
Mecka AI Nears $500M Valuation in Sequoia-Led Round
Mecka AI is closing a Sequoia Capital-led round at roughly $500 million, just three months after its $60 million raise, as investor demand surges for startups supplying human-motion training data to r
Fields Medalists vs. OpenAI: Math Feud Escalates
Twenty-five Fields Medal-winning mathematicians signed an open letter accusing AI labs of threatening their intellectual work, escalating a feud after NYU professor Tristan Buckmaster accused OpenAI o
Google Closes $1.5B+ AI Talent Deal With Mechanize
Google has completed a talent deal reportedly worth over $1.5 billion with San Francisco-based AI coding startup Mechanize Inc., with public LinkedIn profile changes appearing to confirm the transfer
Moonshot AI Targets $2B Revenue Amid Distillation Claims
Moonshot AI is targeting $2 billion in annualized revenue by year-end on the strength of its open-weight K3 model, even as Anthropic publicly accuses the Chinese lab of a massive model distillation ca
Meta Sued Over Use of User Photos to Train AI
Parents in Illinois and California sued Meta in federal court, alleging the company harvested their Facebook and Instagram photos without consent to build its face-recognition system and train generat
Central Eurasia names its 2026 Road to Battlefield winners: Cerberus, WeGlobal AI, and LOOQ
Three AI startups — Cerberus (Uzbekistan), WeGlobal AI (Kazakhstan), and LOOQ (Japan) — won the Central Eurasia regional final of Road to TechCrunch Startup Battlefield 2026 and will compete at TechCr
Top Mathematicians Outraged by OpenAI Methods
Top mathematicians have publicly expressed outrage at OpenAI's methods, according to a new report in The Economist.
Nscale Adds Ex-OpenAI Exec Simo to Board
U.K. AI data center startup Nscale has appointed former OpenAI executive Fidji Simo to its board, joining Sheryl Sandberg and Nick Clegg, as the company prepares for a fall IPO and a reported $3.5 bil
Anthropic Says Seven China-Based AI Labs Ran Industrial-Scale Claude Distillation Attacks
Anthropic exposed industrial-scale AI distillation attacks by seven China-based labs, including Alibaba and Moonshot, which covertly extracted Claude’s capabilities using fake accounts and proxy netwo
Anthropic Details Four AI Hacking Incidents in Report
Anthropic disclosed four incidents this year in which its AI models hacked external companies, the same week a former OpenAI researcher publicly resigned from Anthropic warning that frontier AI could
AI is both the best and worst thing to ever happen to mathematics
OpenAI's AI agents have solved a 200-year-old Millennium Prize Problem — proving the Navier-Stokes equations can blow up — in just 88 hours, triggering a reckoning over whether mathematics can survive
Anthropic: Claude Used in Autonomous Cyber Attacks
Anthropic's 154-page report documents how its Claude AI models were misused by state-sponsored hackers, cybercriminals, and propaganda operators between December 2025 and August 2026 for cyber attacks
DLSS Multi Frame Generation Has Now Been Modded Onto RTX 40, RTX 30 & Even RTX 20 GPUs - Wccftech
Modders have ported NVIDIA's DLSS Multi Frame Generation onto RTX 40, RTX 30, and even RTX 20 series GPUs via unofficial mods, while a separate effort ran DLSS 5 on the NVFP4 format with only a 2% gai
Anthropic Disrupts Russian Hackers' AI Malware Workflow
Anthropic disrupted a Russian state-sponsored campaign (GTG-20006, linked to APT29) that used Claude to autonomously rebuild malware after security detection, hitting 20+ government and defense organi
Show HN: Clawfight.ai MCP-driven agentic game play
Clawfight.ai is an MCP-driven experiment where AI agents rap or fight each other under the pressure of human spectators, with the creator teasing the project on a Show HN post.
Get the Tech newsletter
Curated tech stories, every morning. Free.
No spam. Unsubscribe anytime.