Large language models (LLMs) have exhibited remarkable capabilities across various domains. The ability to call external tools further expands their capability to handle real-world tasks. However, LLMs often follow an opaque reasoning process, which limits their usefulness in high-stakes domains where solutions need to be trustworthy to end users. LLMs can choose solutions that are unreliable and difficult to troubleshoot, even if better options are available. We address this issue by forcing LLMs to use external -- more reliable -- tools to solve problems when possible. We present a framework built on the tool-calling capabilities of existing LLMs to enable them to select the most reliable and easy-to-troubleshoot solution path, which may involve multiple sequential tool calls. We refer to this framework as LIT (LLMs with Inspectable Tools). In order to support LIT, we introduce a new and challenging benchmark dataset of 1,300 questions and a customizable set of reliability cost functions associated with a collection of specialized tools. These cost functions summarize how reliable each tool is and how easy it is to troubleshoot. For instance, a calculator is reliable across domains, whereas a linear prediction model is not reliable if there is distribution shift, but it is easy to troubleshoot. A tool that constructs a random forest is neither reliable nor easy to troubleshoot. These tools interact with the Harvard USPTO Patent Dataset and a new dataset of NeurIPS 2023 papers to solve mathematical, coding, and modeling problems of varying difficulty levels. We demonstrate that LLMs can achieve more reliable and informed problem-solving while maintaining task performance using our framework.

But applied correctly it can make IT infrastructure disappear. But Ronak Chokshi, who leads product marketing for InfoSight at HPE, argues that when considering how to better manage their infrastructure, tech leaders need to consider what services like Uber or Google Maps have achieved. The IT infrastructure behind the delivery of these services is immaterial to the rest of the world – except perhaps for frazzled tech leaders in other sectors who wonder how they could achieve similarly seamless operations. "The consumers don't really care how it works, as long as the service is available when needed, and it's easy to manage," he says. Or, to put another way, says Chokshi, InfoSight worries about the infrastructure, so tech teams can be more application-centric.

But applied correctly it can make IT infrastructure disappear. But Ronak Chokshi, who leads product marketing for InfoSight at HPE, argues that when considering how to better manage their infrastructure, tech leaders need to consider what services like Uber or Google Maps have achieved. The IT infrastructure behind the delivery of these services is immaterial to the rest of the world - except perhaps for frazzled tech leaders in other sectors who wonder how they could achieve similarly seamless operations. "The consumers don't really care how it works, as long as the service is available when needed, and it's easy to manage," he says. Or, to put another way, says Chokshi, InfoSight worries about the infrastructure, so tech teams can be more application-centric.

Your Google Home or Google Assistant speaker listens intently to your every word to deliver music requests, smart home control, and answers to questions -- but only if it's connected to the internet.

As key functions in our society become more digitized, we come to rely on artificial intelligence (AI) to run these processes. While AI and its specialized forms such as machine learning (ML) are powerful tools, its practitioners generally lack insight into how these tools generate outputs. The lack of explainability and observability prevents AI and ML from being trusted tools in critical areas of human activity, such as getting approved for a loan or undergoing procedures to diagnose a disease. Fortunately, Jason Lopatecki and Aparna Dhinakaran understand the importance of explainability and observability of AI and ML in their prior professional experiences, creating Arize AI as the tool they never had. Arize AI is production-grade infrastructure tool used by developers to monitor, assess, understand how their deployed AI works.

Know Your Stuff is a new column that unlocks the hidden secrets about the everyday products you own. You've finished opening gifts and now you're the proud owner of a new…gadget. Maybe you asked for it specifically or maybe it was just a well-intended notion. Either way, now it's yours, it looks expensive, and you don't have a clue how to use it. This is the point at which many of us throw in the towel.

"Nokia's AI driven access analytics solution has given us the ability to proactively address issues, reducing customer calls by solving multiple issues in a single intervention and creating overall efficiencies in our troubleshooting process." High-resolution video, cloud services and the multiplication of connected devices require higher bandwidth as well as increased reliability. Today, for the copper medium to remain competitive relative to fiber, a similar quality of experience is expected from both an end-user and maintenance perspective. This is especially true given the uptake of Fiber in Europe, which stands at less than 50% for home subscribers. As a Tier-1 European service provider, POST Luxembourg was looking to improve the overall performance of its copper troubleshooting process to reduce OPEX and improve satisfaction among its customers.

In this blogpost, you will learn how Azure SQL Database intelligent performance feature Intelligent Insights has successfully helped a customer troubleshoot a hard to find 6-month intermittent database performance issue in a single day only. You will find out how Intelligent Insights helps an ISV operate 60,000 databases by identifying related performance issues across their database fleet. You will also learn how Intelligent Insights helped an enterprise seamlessly identify a hard to troubleshoot performance degradation issue on a large-scale 35TB database fleet. Azure SQL Database, the most intelligent cloud database, is empowering small and medium size business, and large enterprises to focus on writing awesome applications while entrusting Azure to autonomously take care of running, scaling, and maintain a peak performance with a minimum of human interaction, or advanced technical skill set required. Intelligent Insights is a new disruptive intelligent performance technology leveraging the power of artificial intelligence (AI) to continuously monitor and troubleshoot Azure SQL Database performance issues with a pinpoint accuracy and at a large scale simply not possible before.

The days of manually troubleshooting networking problems may soon be over if the announcements today from Cisco and Mist are any indication. The two vendors today each debuted new offerings that employ artificial intelligence technologies to simplify the management of Wi-Fi and wired networks. Cisco Systems used an event in Barcelona, Spain to introduce what it dubs its "second wave of intent-based networking innovation, that includes new "assurance technologies." Specifically, the company is rolling out the Cisco Network Assurance Engine, which uses mathematical modeling and predictive analytics to accomplish a range of tasks, including predicting the impact of changes; continuously verifying network behavior; and checking for security and compliance risks. "Today, IT teams spend 43 percent of their time troubleshooting network issues," Cisco's Scott Harrell says in a blog post. "They are challenged with finding the proverbial needle in a haystack.