Subscribe
Share
Share
Embed
Your Daily Dose of Artificial Intelligence
π§ From breakthroughs in machine learning to the latest AI tools transforming our world, AI Daily gives you quick, insightful updatesβevery single day. Whether you're a founder, developer, or just AI-curious, we break down the news and trends you actually need to know.
Welcome to Daily Inference, your daily briefing on the world of artificial intelligence. I'm your host, and today is March 13th, 2026. We've got a packed episode covering everything from ancient materials finding new life in cutting-edge chips, to AI drawing battle lines with the Pentagon. Let's get into it.
Before we dive in, a quick word from today's sponsor, 60sec.site. Need a website fast? 60sec.site uses AI to build you a stunning, professional website in literally sixty seconds. No coding, no hassle. Check them out at 60sec.site.
Alright, let's start with a story that sounds like it's straight out of a materials science time machine. A South Korean company called Absolics is this year beginning commercial production of glass-based panels designed specifically for next-generation AI chips. Yes, glass β the same material humans have been making for thousands of years β is now poised to become a foundational building block inside the most powerful data centers on the planet. The idea here is that glass substrates can handle the thermal and electrical demands of advanced processors better than traditional materials, potentially unlocking more powerful and efficient AI hardware. This is a fascinating case of old-world technology meeting bleeding-edge computing needs, and it signals that the chip supply chain is about to get a whole lot more interesting as companies search every corner of materials science for that next performance leap. And speaking of chips, Nvidia's Jensen Huang is taking the stage at GTC 2026 β the GPU Technology Conference β to lay out his vision for the future of computing and AI. Expect big announcements and an even bigger vision from the company that has arguably done more than any other to fuel the current AI boom.
Now let's talk about a story that has Silicon Valley, the Pentagon, and the entire AI ethics world on edge. Anthropic, the maker of the Claude AI assistant, finds itself in a high-stakes legal confrontation with the Department of Defense. The Pentagon designated Anthropic as a so-called supply chain risk β a designation typically reserved for foreign adversaries β effectively threatening to bar the company from government work. Anthropic pushed back hard, filing a lawsuit arguing its First and Fifth Amendment rights were being violated. Here's where it gets really interesting: the root of the dispute isn't just about contracts. Anthropic drew two firm red lines. One involved autonomous weapons systems. The other was mass surveillance. Anthropic's leadership was deeply concerned that agreeing to vague language about 'all lawful uses' would, given the NSA's long history of creatively reinterpreting what words like 'target' actually mean, open the door to using Claude for sweeping surveillance of American citizens. Microsoft, Google, Amazon, Apple, and even OpenAI have all filed briefs supporting Anthropic's legal challenge, creating a remarkable moment where the entire AI industry is essentially telling the Pentagon: there are limits. The broader implication here is enormous β we are watching in real time as the AI industry attempts to define the ethical guardrails of its technology against a government that has historically found clever workarounds to expand its surveillance reach.
On a related note about AI accountability, a Tennessee grandmother named Angela Lipps spent nearly six months in jail after a facial recognition AI system incorrectly linked her to a bank fraud case in North Dakota β a state she says she has never even visited. Her story is one of the clearest and most painful illustrations of what happens when AI systems with known accuracy problems, particularly around misidentifying certain demographic groups, are used to make high-stakes decisions without adequate human oversight. At the same time, the UK's anti-fraud body Cifas reported that AI-fueled scams drove fraud reports to a record 444,000 cases last year, with criminals using AI to conduct what they describe as industrialized deception at a scale never seen before. And in a separate but connected development, Truecaller β the caller identification platform with over 450 million users β just launched a new family protection feature that lets one person manage fraud call alerts and even hang up on behalf of vulnerable family members. These three stories together paint a vivid picture of AI as both the tool of the scammer and, increasingly, the shield against them.
Let's shift to something a bit more hopeful. Google's AI research team just released a project called Groundsource, which uses their Gemini model to do something genuinely clever: it scans through unstructured historical news reports β the kind of messy, narrative text that databases have never been able to make sense of β and converts it into clean, structured, usable data. The first major output is a dataset of 2.6 million historical urban flash flood events across more than 150 countries. Why does that matter? Because when it comes to rapid-onset natural disasters like flash floods, there's a massive gap in historical data that emergency planners and climate scientists desperately need. This is AI as a kind of archaeologist for information β digging useful facts out of archives that would otherwise take generations of human researchers to process. On a related note, Google also rolled out its new 'Ask Maps' feature powered by Gemini, which lets you ask Google Maps genuinely complex, real-world questions and plan trips conversationally. And separately, on Samsung's S26 Ultra, Gemini's task automation feature just went live in beta β meaning your phone can now literally use apps on your behalf, ordering food or booking rides while you watch. The future of the AI assistant is no longer just answering questions. It's doing things.
Finally, the AI jobs-and-economy story continues to evolve in striking ways. Atlassian, the software collaboration giant, laid off around 1,600 workers β roughly 10 percent of its workforce β explicitly to redirect resources toward AI development. This follows similar moves by Block and others. At the same time, AI coding platform Replit just hit a 9 billion dollar valuation, tripling in just six months, while vibe-coding startup Lovable added 100 million dollars in revenue in a single month with only 146 employees. Sales automation startup Rox AI hit a 1.2 billion dollar valuation despite being founded just two years ago. And Gumloop landed 50 million dollars from Benchmark to help every employee become an AI agent builder. The message from investors is clear: the companies that help organizations actually deploy and use AI β not just build the underlying models β are where the money is flowing right now.
That is your Daily Inference for March 13th, 2026. Whether it's glass chips or Pentagon showdowns, AI is reshaping every corner of our world at a pace that's genuinely hard to keep up with β which is exactly why we're here. For deeper dives on every story we covered today, head to dailyinference.com and sign up for our daily AI newsletter. We break it all down for you every morning. And again, huge thanks to today's sponsor, 60sec.site β build your AI-powered website in sixty seconds flat. Until tomorrow, stay curious, stay informed, and keep inferring.