Data science
fromComputerworld
3 days agoAI project 'failure' has little to do with AI
The reliability of genAI is compromised by various factors, necessitating independent verification of its outputs.
In the nineteenth century, entire railway networks became obsolete almost overnight, not due to physical deterioration, but because of changes in the technical standards that supported them. The expansion of railroads across Europe and North America adopted different track gauges, and as a dominant standard gradually emerged, these infrastructures became incompatible with one another.
Perforated metal has long been valued for its strength, versatility, and clean visual appeal. Created by punching patterns of holes into metal sheets, it offers a practical balance between airflow, light control, and structural support. Across industries such as architecture, construction, mining, and interior design, perforated metal has become a go-to material for projects that require both function and style.
In enterprise commerce, totals don't drift because someone forgot algebra. They drift because reality changes: promos expire, eligibility changes when an address arrives, catalog data updates, substitutions happen, and returns unwind prior discounts. When someone asks "why did the total change?" you need more than narration. You need evidence - a trail of facts you can replay and a pure computation that deterministically produces the same result.
The blank canvas wasn't a hurdle; it was an invitation. An invitation to think, to wrestle, to connect disparate dots until a clear, compelling strategy emerged. Today, that invitation often comes in the form of a blinking cursor in a prompt box. The promise is seductive: speed, efficiency, and democratized creativity.
Some 337,000 cars, 29,000 of them in Germany, covering five different models are "potentially concerned" by the safety issue, which concerns incorrect routing of the dashboard wiring, said the KBA. The recall concerns the i5, 5, M5, i7 and 7 models built between June 2022 and December 2025.
the automation of heavy machinery enabled plants to operate continuously, increasing productivity and revenue. The downside was that any small hiccup was acutely felt, cascading through the production line. At first, it was assumed that inadequate lubrication of factory equipment was causing parts to seize up or break apart. And so, the Lubrication and Wear Group of the Institution of Mechanical Engineers, along with the Iron
Siemens has published eight new advisories. The company has released patches and mitigations for high-severity issues in Desigo CC, Sentron Powermanager, Simcenter Femap and Nastran, NX, Sinec NMS, Solid Edge, and Polarion products. A medium-severity flaw has been found in Siveillance Video Management Servers. Exploitation of the vulnerabilities can lead to unauthorized access, XSS, DoS, code execution, and privilege escalation.
French startup WheelMove demonstrated a compact motorized wheelchair add-on at CES 2026 that not only acts as a power-assist device, but can also lift up a chair's small front wheels so it can ride over rough terrain. The accessory upgrades a manual wheelchair with a 10-inch extra wheel and a 250W motor that can drive at up to six miles per hour (10 kilometers per hour) and handle slopes up to 10 percent, with a range of 15.5 miles (25km).
To find the typical example, just observe an average stand-up meeting. The ones who talk more get all the attention. In her article, software engineer Priyanka Jain tells the story of two colleagues assigned the same task. One posted updates, asked questions, and collaborated loudly. The other stayed silent and shipped clean code. Both delivered. Yet only one was praised as a "great team player."
Military weapons are designed to give commanders an advantage, but that advantage is rarely permanent. Systems that once multiplied combat power can become burdens as threats evolve, environments shift, and missions change.Some weapons begin to demand more protection, maintenance, or political consideration than the value they provide. Here, 24/7 Wall St. is taking a closer look at the weapons that became liability issues instead of force multipliers.
Sometimes I'm rather impressed by the ingenuity of those in the business of making scammy gadgets. Over the years, I've examined a wide range of products, from scam devices that claim to save you money on your power bill to a high-power USB charger filled with an unusual goo-like substance. Also: Want to cut your electric bill? Skip these scam 'power-saving' devices - and buy this instead
This extends to the software development community, which is seeing a near-ubiquitous presence of AI-coding assistants as teams face pressures to generate more output in less time. While the huge spike in efficiencies greatly helps them, these teams too often fail to incorporate adequate safety controls and practices into AI deployments. The resulting risks leave their organizations exposed, and developers will struggle to backtrack in tracing and identifying where - and how - a security gap occurred.
We are now in a time of manufacturing where precision is more than a technical necessity; it's a business requirement. The more complex, globally dispersed and demanding things get, the less slack remains in the system. Under these circumstances tolerance management has become a decisive competence and affects competitiveness not only in terms of controlling costs, ensuring quality and improving production efficiency but also for long term market success.
A secure software development life cycle means baking security into plan, design, build, test, and maintenance, rather than sprinkling it on at the end, Sara Martinez said in her talk Ensuring Software Security at Online TestConf. Testers aren't bug finders but early defenders, building security and quality in from the first sprint. Culture first, automation second, continuous testing and monitoring all the way; that's how you make security a habit instead of a fire drill, she argued.
The normative form for interacting with what we think of as "AI" is something like this: there's a chat you type a question you wait for a few seconds you start seeing an answer. you start reading it you read or scan some more tens of seconds longer, while the rest of the response appears you maybe study the response in more detail you respond the loop continues
As they headed toward Highway 24 from Camino Pablo, their child was strapped into her car seat, looking sleepy. Then, in a fraction of a second, something frightening happened. As Shah turned the car onto the horseshoe curve of the ramp, he said, the back passenger door next to their baby swung open violently, all 60 pounds of it, much of it "ultra-hard" stainless steel. Shah and his wife screamed out in shock, and Shah immediately pulled onto the shoulder to see what had happened.
My grandmother's refrigerator ran for forty years. The washing machine she bought in the 1970s? Still spinning when she passed away. Meanwhile, I'm on my third coffee maker in five years, and don't get me started on the laptop that mysteriously died two weeks after the warranty expired. This isn't just bad luck or nostalgia talking. There's something fundamentally different about how products are made today versus decades ago.
When I work on something, whether it's at Interfere or my personal projects, I like to experiment a lot. Design engineering is a lot about trial and error, and I often spend hours trying to find the "this feels right" moment. This is where AI helps. Instead of spending hours on a concept that I'm unsure of, I try that concept out in a matter of minutes, and throw it away if it doesn't feel right.