AI News Roundup – January 16, 2026
Google DeepMind CEO claims China trails Western AI by only months
Google DeepMind CEO Demis Hassabis states that China lags behind Western AI capabilities by only a few months despite ongoing US export restrictions on advanced hardware
dataconomy
Jan 16, 2026
The Real AI Talent War Is for Plumbers and Electricians
The rapid expansion of AI data centers faces a critical bottleneck due to a severe shortage of skilled tradespeople like electricians and plumbers needed to build the necessary infrastructure
wired
Jan 15, 2026
Guiding Principles of Good AI Practice in Drug Development
US and European regulators have established ten guiding principles for human-centric AI use in drug development to ensure patient safety and data reliability while accelerating innovation
u.s. fda
Jan 14, 2026
Chinese AI firm trains state-of-the-art model entirely on Huawei chips
Zhipu AI successfully trained a competitive image generation model using only domestic Huawei chips, proving Chinese self-sufficiency despite US hardware export restrictions
computerworld
Jan 14, 2026
NVIDIA and Eli Lilly Launch $1bn AI Drug Discovery Lab
NVIDIA and Eli Lilly are investing $1 billion in a joint Bay Area lab that uses robotics and physical AI to build biological foundation models for faster drug discovery
aimagazine
Jan 14, 2026
Inside CEO Mustafa Suleyman’s Vision for Microsoft AI
Microsoft AI CEO Mustafa Suleyman focuses on ‘Human Superintelligence’ to build controllable, personalized AI companions designed to solve real-world problems and serve humanity
businesschief
Jan 14, 2026
How AI Companies Got Caught Up in US Military Efforts
Leading tech firms are increasingly pursuing defense contracts due to high infrastructure costs, driving the ‘militarization’ of the AI sector
wired
Jan 14, 2026
Google’s Gemini AI will use what it knows about you from Gmail, Search, and YouTube
Google launches ‘Personal Intelligence’ for Gemini, an opt-in beta allowing the AI to pull context from Gmail, Photos, and YouTube to provide hyper-personalized responses
theverge
Jan 14, 2026
This new, dead simple prompt technique boosts accuracy on LLMs by up to 76% on non-reasoning tasks
A new, straightforward prompting technique reportedly increases large language model (LLM) accuracy by up to 76% on non-reasoning tasks, offering a significant performance boost without complex engineering
venturebeat
Jan 13, 2026
Mark Zuckerberg says Meta is launching its own AI infrastructure initiative
Meta launches the ‘Meta Compute’ initiative to build gigawatts of energy capacity and custom silicon infrastructure to support the development of leading AI models
techcrunch
Jan 12, 2026
AI Talks 