Author: Blaze Woodard
Blaze Woodard, an editor at cubox-i.com, is presently working as an intern at a Silicon Valley technology company while majoring in politics at the University of Kansas. Blaze, who identifies as both a policy thinker and a self-described tech geek, offers a viewpoint on hardware and computing coverage that few editors in this field can match: the capacity to relate the workings of a circuit board to the larger political, regulatory, and social forces influencing the technology sector. Even though her academic path led her to political science, her early fascination with technology persisted. She writes about computing, AI, and hardware with the zeal of someone who truly loves the subject, not as someone assigned to cover it. Blaze plays soccer and spends her free time with friends and living her life, which is exactly what a college student should do outside of the office and newsroom.
It seems almost silly for a 40-gram drone to be hunting mosquitoes in the air. Still, it’s hard not to think that something real is going on when you see the video from Tornyol Systems of a small self-driving drone going after a moth in what the company calls its first confirmed air-to-air kill. Not because getting rid of mosquitoes is a simple goal. It’s not. More than 700,000 people are killed by mosquitoes every year, which is more than all the people killed in all current wars put together. Over 700 million people also get diseases from mosquitoes every…
People in engineering communities build a certain kind of trust over time. It’s not from press releases or demos at large events. It comes from a device that worked fine on a factory floor for two winters. That kind of trust has been earned by the CuBox-i, and for developers across the US building industrial IoT apps, it has quietly gone from being an interesting option to their first choice. The CuBox-i is small, about the size of a Rubik’s cube, and that’s part of its appeal. In some industrial settings, there isn’t always rack space or climate-controlled enclosures. When…
When you walk into a normal data center, the screens and servers are not the first thing that you see. Why is there so much noise? It’s because of a wall of industrial fans that are working nonstop to keep millions of dollars worth of computer hardware from melting. For many years, that noise was just how things worked. It seems like an issue that can’t be set aside any longer. As AI workloads get more complex and dense, the heat loads inside modern data centers are getting too high for air cooling to handle. Clusters of high-performance computers, GPU…
People who are accused of something they didn’t do feel a certain kind of fear. You are now seventeen years old and in the principal’s office. You are told that software has found that your essay was written by a computer. Your paper. The one you stayed on for three nights. This takes place more often than most school administrators want to admit in schools across the country. Since ChatGPT became famous in late 2022, teachers have been nervous, which is understandable. In less than thirty seconds, the tool can make a good five-paragraph essay. Schools did want a defense,…
It is said that there is a warehouse in China full of brand-new, unopened Nvidia H200 servers still in their original packaging. Videos of this warehouse have been shared on social media as if they were a product launch. Nothing for the press. There is no authorized distributor. The only sounds are boxes and the sound of an underground supply chain that no longer needs to be hidden. That picture, which is disturbing in its casualness, says more about how U.S. controls on technology exports are right now than most policy briefings ever could. The U.S. government has spent years…
At every defense tech conference, the same words will be said over and over again: Edge AI, self-driving cars, real-time inference, and combining sensors. The slides look good. The demos are really great. And very little of it shows what really happens when you take that hardware outside and put it somewhere really tough. We now know how to run AI in a lab with controlled temperatures and humidity. That’s a whole different conversation when it comes to running AI on a platform that is moving over rocks, baking in the desert heat, and drawing power from a source that…
When I looked at the specs for a new laptop a few months ago, there it was: “NPU: 45 TOPS.” It made sense to use the acronym. It wasn’t clear what it meant in everyday life. It was also thought, and later proven, that almost none of the software running on that computer was touching the NPU. That’s what really happened in 2026. The hardware has been sent. The software is still taking a deep breath. Neural Processing Units are now built into almost all new smartphones, laptops, and Macs that come out today. They are built into the System…
Collin Burns, a researcher, moved to Washington, D.C., last week for a job that doesn’t exist anymore, at least not for him. He quit Anthropic, packed his things, and started the onboarding process at the Center for AI Standards and Innovation in the Commerce Department. After four days, he was fired. It turned out that the reason was the company he had left to get the job. Burns had worked at OpenAI and Anthropic before, which are two of the best AI labs in the country. In a different time, that resume would have been seen as a strong qualification…
Having that small of a computer in your hand feels almost strange. SolidRun’s CuBox-M is only two inches on each side, which is about the same size as a Rubik’s Cube. The company seems to like comparing its products to puzzles. Inside that plastic case is an NXP i.MX 8M Plus processor with four Arm Cortex-A53 cores running at up to 1.8GHz. It also has what SolidRun calls a “integrated neural processing unit” that can do about 2.3 trillion operations per second. There’s a lot of power in something that you might lose in a desk drawer. The CuBox-M, which…
In the semiconductor industry, progress is often measured by things that are easy to see, like faster GPUs, denser chips, and bigger clusters. Memory is talked about, but it’s not usually the main topic. This is one reason why the announcement of the Rambus SOCAMM2 chipset is more important than it might seem at first. The chipset, which was announced in April 2026, is based on SOCAMM2, which stands for “Small Outline Compression Attached Memory Module 2.” SOCAMM2 is a JEDEC-standard module format that is meant to bring LPDDR5X memory into AI server environments. The speeds are good, with up…