Business
Deloitte Warns AI Threatens Billable Hour Model
Deloitte's internal report projects AI will dominate consulting by 2035, shrinking the billable hour model.
LATEST NEWS
-
AMD Optimizes LLM Inference with Low-Latency GEMMs on GPUs
AMD's new low-latency GEMM optimization improves inference speed for large language models by 1.64x on a K = 7168 decode grid.
-
Claude Code Executes Hidden Malware in GitHub Repos Without Verification
Security researchers found attackers can compromise developers' machines by embedding malware in GitHub repos, bypassing AI tools like Claude Code.
-
Meta Unveils Brain2Qwerty v2 for Non-Invasive Brain-to-Text Communication
Meta has released Brain2Qwerty v2, an AI system achieving 61% word accuracy in decoding non-invasive brain signals into text, surpassing previous non-surgical methods.
-
Samsung and SK Hynix Plan 590 Billion Chip Investment Amid AI-Driven Memory Price Surge
Samsung and SK Hynix, supported by South Korea, plan a $590 billion investment to expand chip production as AI demand drives memory prices up.
-
Flexion Robotics Develops AI for Autonomous Humanoid Tasks
Flexion Robotics has created an AI system that enables humanoid robots to perform complex tasks autonomously, using simulation and reinforcement learning to carry out chores like retrieving parcels and organizing items.
-
HP Expands OpenAI Frontier Partnership for Enterprise AI Integration
HP Inc. announced June 28, 2026, it will scale its OpenAI Frontier partnership after successful pilots, with AI tools improving productivity and customer experiences.
-
Ford Rehires Veteran Engineers After AI Fails to Meet Quality Goals
Ford has rehired 350 veteran engineers after AI-driven quality systems underperformed, aiming to cut $1 billion in costs this year.
-
Micron Surpasses Meta and Tesla in Market Cap
Micron's market cap reached $1.27 trillion, briefly overtaking Meta and Tesla, driven by AI demand for memory chips.
-
CEO-Bench Test Reveals Only Three AI Models Surpass Startup Capital
In a 500-day startup survival simulation, only three AI models finished with more than the initial $1 million in capital, according to a Princeton University study.
-
Tencent Researchers Argue AI Must Finish Tasks, Not Just Answer Questions
A new survey paper by Tencent's Youtu Lab and Chinese universities argues AI systems must shift from generating answers to completing tasks reliably, with a focus on reusable skills and persistent work environments.
-
Sina's VibeThinker-3B Model Matches Top Models on Math and Coding
Sina's VibeThinker-3B, a 3-billion-parameter model, matches top models up to 333 times its size on math and coding benchmarks, according to a technical report.