Latest AI (Artificial Intelligence) News
NVIDIA advances âphysical AIâ for autonomous vehicles with Alpamayo platform
NVIDIA introduced **Alpamayo**, a new autonomous driving platform built around *Alpamayo 1*, a 10âbillionâparameter visionâlanguageâaction model that uses chainâofâthought reasoning to navigate complex driving scenarios. The system is designed to improve realâtime decisionâmaking and safety in selfâdriving cars, signaling a broader push toward physical AI systems that interact with the real world.
TII unveils FalconâH1R, a compact 7B reasoning model rivaling much larger systems
The UAEâs Technology Innovation Institute launched **FalconâH1R 7B**, a TransformerâMamba hybrid model that delivers reasoning performance comparable to models up to seven times larger. Its efficiency and small footprint target onâdevice and resourceâconstrained deployments, reflecting a wider industry move toward specialized, smaller AI models instead of everâlarger LLMs.
Agentic AI market forecast to soar from $5.2B in 2024 to $200B by 2034
New industry analysis projects the **agentic AI** marketâsystems that can autonomously execute multiâstep tasksâto grow from $5.2 billion in 2024 to **$200 billion by 2034**. Growth is driven by demand for smaller, taskâspecific agents embedded in business workflows, replacing simple chatbots with operational tools that plan, act, and coordinate across applications.
DrugCLIP AI platform screens 10,000 proteins and 500M compounds in a day
Researchers at Tsinghua and Peking University released **DrugCLIP**, an AI screening method that can virtually evaluate 10,000 human proteins against 500 million compounds in about one day. The system generated roughly 2 million promising smallâmolecule âhitsâ and identified ligands for clinically relevant targets, potentially compressing years of earlyâstage drug discovery into days.
Global AI adoption reaches roughly one in six people, with a widening digital divide
Microsoftâs latest **Global AI Adoption Report** finds that generative AI usage continued to rise in late 2025, with about **one in six people worldwide** now using such tools. The report highlights a widening divide: the U.S. leads in infrastructure and model development but ranks only 24th in usage, while smaller highly digitized economies and countries like South Korea show much higher adoption.
Chinese openâsource platform DeepSeek rapidly expands across Africa and sanctioned markets
The report also notes the rapid rise of **DeepSeek**, an openâsource Chinese AI platform released under an MIT license with a free chatbot that lowers financial and technical barriers. DeepSeek is gaining strong traction in China, Russia, Iran, Cuba, Belarus, and across Africa, making AI access a new front in geopolitical competition between U.S. and Chinese ecosystems.
AI model watermarking market expected to triple to $1.17B by 2029
A new global market report projects **AI model watermarking**âtechnologies that embed invisible identifiers into AI outputsâto grow from $0.42 billion in 2025 to **$1.17 billion by 2029**. Demand is fueled by concerns over deepfakes and misinformation, regulatory pushes for AI transparency, and the need to protect intellectual property as generative models proliferate.
U.S. Department of Energy commits over $320M to AIâdriven scientific machine learning
The U.S. Department of Energy announced **more than $320 million** in investments to accelerate AI capabilities for scientific discovery as part of its Genesis Mission. Funding includes support for the new LEADS Institute, which will develop scientific machine learning algorithms for largeâscale data, digital twins, and realâtime information extraction across multiple scientific domains.
Financial institutions face AI compliance as a âregulatory necessityâ in 2026
Analysts report that for financial institutions, AI is shifting from an optional efficiency tool to a **regulatory necessity** in compliance workflows. Firms are expected to use AI for monitoring, risk assessment, and reporting while simultaneously preparing for tighter oversight on model governance, explainability, and responsible AI use.
Unisys predicts three enterprise AI applications will dominate highâROI deployments
A new Unisys report forecasts that **chatbots**, **AI coding agents**, and **AIâdriven service assistants** will emerge as the most repeatable, highâROI AI applications in enterprises. The company expects 2026 to emphasize quality and measurable outcomes over costâcutting, with organizations favoring smaller, domainâspecific models and using AI both to enhance cyber defenses and to counter increasingly AIâenabled attacks.
Construction industry moves from AI pilots to full workflow integration
Experts in construction technology say 2026 marks a shift from experimental AI pilots to **deep integration** in core construction workflows. Machine learning and generative AI are being used for document processing, clash detection, schedule optimization, realâtime safety monitoring via computer vision, and AI âcoâworkerâ assistants that track tasks and generate reports.
Biotech sector leans on AI for drug development and humanârelevant testing models
Biotech analysts highlight 2026 as an inflection point where **AIâenabled drug development** and humanârelevant testing models move from early discovery into clinical validation and regulatory review. AI is enabling more precise target identification, better experimental design, and integration of complex multiâomic and spatial biology data, reshaping how labs operate day to day.