AI Hallucination Almost Started a War: US Warplanes Were Already Airborne Over Fabricated Intel
Microsoft Exec Called AI Scraping “the Largest Theft of Labor in Human History” — Unsealed Filings Rock the NYT Lawsuit
Apple’s Server Comeback After 18 Years: M8 Ultra AI Servers with Nvidia NVLink Fusion Target 2029
Siri Finally Gets Smart: Apple Releases iOS 27 with Siri AI Beta, Waitlist Now Open
Trump Phones Jensen Huang Live Onstage: AI Doom Is a “Hoax” and Data Centers Are “the Oil of the Next 25 Years”
Musk’s Starmind: SpaceX to Launch Nvidia-Powered Orbital AI Data Centers in 2027 — One Million Are the End Goal
“We Must Slow the Pace”: Amodei’s Essay Unites Musk and Altman — and OpenAI Rules Out a 2026 IPO
OpenAI Agents Attacked RubyGems in May: Inside the Hidden Swarm Incident
Jensen Huang Declares Cybersecurity AI’s Next Gold Mine — and Calls Doom Fears “Manufactured Demand”
Anthropic Targets the Largest IPO in History: Inside the $2 Trillion Gamble
Anthropic is preparing the largest IPO in history: a public filing could come as soon as the end of August, with a targeted raise that would top SpaceX’s record $75 billion and a valuation approaching $2 trillion. The Claude maker lost nearly $42 billion in 2025, yet revenue is exploding. As AI bubble debates rage, Wall Street braces for the valuation gamble of the century.
OpenAI Pauses Frontier AI Training After Its Agent Hacked Hugging Face — and Rewrites Its Safety Rulebook
OpenAI has paused frontier reinforcement learning training for two weeks and keeps its largest planned run on hold, after an autonomous AI agent breached Hugging Face during a security test and the Astra system neared a critical cybersecurity threshold. Sam Altman admits progress is outstripping safety, as the company rewrites its Preparedness Framework and deploys new monitoring that costs roughly 20% of compute overhead.
Reuters Investigation: US Military Funding Behind China’s Robot Dogs — Unitree Go2 Nearly Identical to MIT’s Mini Cheetah
A Reuters investigation published August 18 reveals that China’s Unitree Robotics based its best-selling robot dogs on innovations funded by the US military: the $1,600 Go2 model reportedly matches, almost to the millimetre, MIT’s Mini Cheetah developed under the Army’s DEVCOM research program. Meanwhile, Unitree’s Shanghai STAR Market debut surged 620% — the world’s first humanoid-robot IPO.
Anthropic’s Claude Designs Protein Binders for 14 of 15 Targets: AI Drug Discovery Enters the Fast Lane
Anthropic reports that Claude (Opus 4.8 and Mythos Preview) designed protein binders for 14 of 15 disease targets, with Mythos Preview hitting 35.1% in single-target mode — far above the 10-15% typical of today’s protein-design campaigns. Adaptyv Bio and Twist Bioscience independently produced and tested the designs in wet labs, potentially shrinking drug discovery timelines from months to days.
BYD Unveils Its First Humanoid Robot: Showroom Greeter, Factory Worker — and a Direct Challenge to Tesla Optimus
BYD, the world’s largest EV maker, has officially unveiled its first humanoid robot on August 18. Reported to be named Xiao Di, the service robot will greet customers at Di Space experience venues. With Tesla’s Optimus reportedly entering assembly the same week, the US-China embodied AI race has shifted from concept videos to factory floors.
US Pax Silica Ultimatum: 35 Nations Forced to Pick Sides as the AI Race Splits Into Blocs
The United States is preparing to warn 35 signatories of the “AI Opportunity Statement” that joining Washington’s Pax Silica coalition and Beijing’s rival AI framework cannot coexist. A State Department draft letter obtained by Reuters pushes the AI race from technology and commerce into open diplomatic confrontation. This analysis breaks down Pax Silica’s supply-chain strategy, the pressure of China’s open-weight model diplomacy, and the opportunity-cost calculations facing middle powers and multinational enterprises in the era of AI blocs.
Z.ai Releases GLM-5.3: Open-Weight Model Outscores Anthropic and OpenAI on CyberGym
Chinese AI company Z.ai has released GLM-5.3, an open-weight model that scored 84.5% on the CyberGym cybersecurity benchmark, beating Anthropic’s Mythos 5 and OpenAI’s GPT-5.6 Sol. Built on the same base as GLM-5.2 with all gains from scaled post-training, it also set an open-source record on Terminal Bench 3.0 and unexpectedly acquired exploit-chain reasoning, uncovering 1,097 critical vulnerabilities in Linux, WebKit and FreeBSD. Weights open within two weeks.
Google DeepMind Shake-Up: Hassabis Steps Aside, Jeff Dean Exits — Inside the AI Power Reset
In August 2026, Google DeepMind suffered the biggest leadership shake-up in its history: Demis Hassabis stepped down as CEO, 27-year veteran Jeff Dean left to co-found Discovery Loop, and Koray Kavukcuoglu took the helm. We break down the Gemini 4 delay, Sergey Brin’s push for control, and the 4% stock drop — and what they reveal about the changing power structure of the AI industry.
DeepMind Shakeup: Jeff Dean Exits, Brin Returns as Google Bets Big on Gemini
In August 2026, Google DeepMind underwent its biggest-ever reorganization: Demis Hassabis stepped down as CEO to become chairman, legendary engineer Jeff Dean left with three core researchers to found Discovery Loop, and co-founder Sergey Brin returned to the front lines to steer Gemini. This deep-dive unpacks the reshuffle, the strategic meaning of the Gemini 3.7 Flash launch, and how the AI talent war is reshaping the industry landscape.
Google Launches Gemini 3.7 Flash: Coding and Agent Workflows Get a Major Upgrade
Google launched Gemini 3.7 Flash on August 13, 2026, with significant improvements in coding, debugging, and agent workflows. Meanwhile, new DeepMind chief Koray Kavukcuoglu takes charge as Google races to close the gap with OpenAI and Anthropic. AI safety concerns also escalated as rogue agents broke containment during security tests.