Hugo Bowne-Anderson and Alan Nichol    TL;DR: Enterprise AI teams are discovering that purely agentic approaches dynamically chaining LLM calls don’t deliver the reliability needed for production systems. The prompt and pray model where business logic lives entirely in prompts creates systems that are unreliable, inefficient, and...

Mike Loukides    Welcome to our annual report on the usage of the O’Reilly learning platform. It’s been an exciting year, dominated by a constant stream of breakthroughs and announcements in AI, and complicated by industry wide layoffs. Generative AI gets better and better but that trend may be at an end. Now the...

Mike Loukides    Despite its days, December is a short month. It’s hard for announcements and happenings other than office parties to get attention. Fighting this trend, OpenAI made a series of announcements: their Days of OpenAI. Not to be upstaged, Google responded with a flurry of announcements,...

Sebastian Hassinger    Making quantum computers is really very difficult. The quantum bits, or qubits, are made up of superconducting circuits operating at thousands of a Kelvin above absolute zero, or individual atoms, or photons. Aside from the challenges of engineering at those extremes, there is the whole matter of...

Mike Loukides    Like just about everyone, we were impressed by the ability of NotebookLM to generate podcasts: Two virtual people holding a discussion. You can give it some links, and it will generate a podcast based on the links. The podcasts were interesting and engaging. But they also had some limitations. The...

Mike Loukides    Alibaba’s latest model, QwQ B Preview, has gained some impressive reviews for its reasoning abilities. Like OpenAI’s GPT o , its training has emphasized reasoning rather than just reproducing language. That seemed like something worth testing out or at least playing around with so when I heard...

Mike Loukides    It’s the end of the year for Radar We hope all of our readers enjoy the holidays. Here’s one prediction for : Is this the end of the road for improving LLM performance by scaling either the number of parameters or the training data No one knows yet. Regardless of the answer, we expect...

Noah Gift    There is a crisis in technical education. The golden road to a career has always been through a college education. However, this golden road has developed deep cracks and is badly in need of maintenance. Postsecondary education is rapidly becoming unaffordable, even at public colleges and...

Paco Nathan    One popular term encountered in generative AI practice is retrieval augmented generation RAG . Reasons for using RAG are clear: large language models LLMs , which are effectively syntax engines, tend to hallucinate by inventing answers from pieces of their training data. The haphazard results...

Mike Loukides    Between vacation, end of year projects, the coming holidays, and other hysteria, I haven’t come up with an article this month. So here’s a quick list of things that have amazed me recently. Are we virtual yet I’m far from the first person to find NotebookLM amazing, and I certainly won’t be the...

Mike Loukides    October had many language model releases. The mid size models, and even the small models, are catching up to frontier models like GPT . o in performance. But the release that blew us all away wasn’t a language model: It was Claude’s computer use API. Computer use allows you to teach Claude how to...

Eric Colson    Modern organizations regard data as a strategic asset that drives efficiency, enhances decision making, and creates new value for customers. Across the organization product management, marketing, operations, finance, and more teams are overflowing with ideas on how data can elevate the business. To...

Trisha Gee    I’m worried about AI. I’m not worried about it taking my job. I believe AI is a genuine productivity tool. By which I mean it can make developers produce more. The question is whether those developers are producing something good or not. The difference between an experienced developer and a junior...

Mike Loukides    Back in August, I cavalierly said that AI couldn’t design a car if it hadn’t seen one first, and I alluded to Henry Ford’s apocryphal statement If I had asked people what they wanted, they would have said faster horses. I’m not backing down on any of that, but the history of technology is always...

Mike Loukides    In August , we asked our customers to tell us about security: their role in security, their certifications, their concerns, and what their companies are doing to address those concerns. We had, complete responses, of which roughly one third are members of a security team. ...