Highlights from MLSys 2026

by Capital One Tech — Capital One’s AI research team recaps MLSys 2026, including optimizing serving LLMs, RAG and agentic AI.

Capital One Tech·1h ago

Machine Perception · Part 5

by bowie fungo — The contract tells the runtime what the world should look like. Memory is how it knows what the world has actually done.

Towards AI·2h ago

Why We Built an OS for AI — and What Taku 1.0 Taught Us

by Taku AI — In 2025, AI could already think, write code, and plan. The harder problem was getting it to actually do things. Taku 1.0 was our first…

Medium·2h ago

NVIDIA Nemotron 3.5 Lightning and NeMo Switchyard Deliver Faster, Smarter, More Efficient Agentic AI

As AI shifts from chatbots to autonomous agents, open models are serving market demands for full control over where AI runs and how it’s deployed and evolves. Today, NVIDIA is expanding its Nemotron...

NVIDIA·4h ago

Can a Local LLM Run My AI Assistant?

by Arsen Apostolov — I replayed the same 27 real production tasks through two local models, one hardware upgrade apart, to find out what it actually takes to replace Claude as the brain behind a 90-tool personal agent....

Towards Data Science·5h ago

The App Trap: Are You Getting Things Done, or Are You Getting Away?

by Mehekk Bassi — What does ‘productivity’ even mean at this time and age? Does it mean getting things done, or creating systems for getting them done?

Voice Out!·2h ago

What Do You Give Someone Who Already Has Everything?

by Peter Anderson — When buying another thing isn’t the answer, the best gifts are often the ones with a story behind them.

Medium·2h ago

Claude vs Claude Cowork: What’s the Difference and Which One Should You Use in 2026?

by Iamnaveediqbalqau — Artificial intelligence is moving quickly.

Medium·2h ago

A New Trick Reveals AI Models’ Inner Thoughts

Researchers devised a way to extract “reasoning traces” from Claude, GPT, and Gemini. What they found, they say, indicates that some Chinese AI may be trained on leading US models.

Wired AI·6h ago

What smart people are saying about Meta's open-weight model Muse Glimmer and Mark Zuckerberg's AI manifesto

Even the White House took notice of Meta CEO Mark Zuckerberg's AI manifesto. Others focused on the release of Meta's open-weight model.

Insider on MSN·7h ago

Beyond the Sandbox: What AISI’s Incident Report Reveals About Agentic AI Security

by Alessandro Pignati — An inside look at how frontier AI agents breached scope during cyber testing, deceived humans, and what it means for enterprise AI…

Medium·2h ago

What Does Your Podcast Library Say About You?

by Alla Khrystych — I treated my Castbox subscriptions as a professional digital footprint. ChatGPT found a pattern I had never explicitly planned.

Medium·2h ago

TAI #217: AI Agents Are Finding Attack Paths We Never Approved

by Towards AI Editorial Team — Also, Meta’s return to open weight with Muse Glimmer and Spark 1.2, DeepMind leadership reshuffle & more!What happened this week in AI by LouieMeta made a welcome return to open weights this week....

Towards AI·3h ago

Novo Nordisk and AWS bring agentic AI into drug discovery

Novo Nordisk is expanding its use of AWS artificial intelligence tools across drug discovery, including AI agents for target identification, therapy design, and research workflows. Under the...

AI News·7h ago

Spotlight on agentic AI | Agentic AI and competition law (via Passle)

Bristows·7h ago