Ai

The conventional reliability engineering toolkit was built for systems that behaved the same way each time given the same input. AI agents do not behave the same way each time given the same input. The classic tools - unit tests, integration tests, deterministic replay, traditional monitoring - all assume a property that the systems being operated do not have. This mismatch is not a small operational annoyance; it is the central challenge of running AI agents in production, and the patterns for handling it are still being worked out. ...

Dario Amodei is one of the few frontier-lab CEOs whose public talking points have not changed materially in five years. The same message he gave to small audiences in 2021 - that powerful AI is coming faster than people think, that the safety problem is real, and that the companies building it have an obligation to do so carefully - is the message he is giving to Congress and Davos in 2026. The thing that has changed is that he now runs the company most aggressively turning that message into a commercial position. ...

AlphaFold’s release in 2021 was the AI-for-science moment that broke through to the general public. A computational solution to a 50-year-old problem in biology - predicting protein structure from sequence - that produced a tool used by hundreds of thousands of researchers. The narrative around AI-for-science crystallised: deep learning would produce a series of similar breakthroughs across scientific domains. The 2026 reality is more interesting and less clean. AlphaFold-class breakthroughs have been rarer than the early narrative suggested. But AI has spread across scientific practice in subtler ways that, in aggregate, have done more to change how science is actually done than the few headline breakthroughs. ...

The open-weight model conversation in 2023 was about whether the open ecosystem could keep up with the frontier labs at all. The conversation in 2024 was about how big the gap was. The conversation in 2026 has changed shape: on most benchmarks that matter to most production workloads, the open-weight ecosystem has either closed or substantially narrowed the gap, and the strategic question is no longer “can we use open models” but “which open model fits this workload best.” ...

For most of the AI conversation in 2024 and 2025, the binding constraints on the build-out were chips and capital. By 2026 the conversation has shifted, and the constraint that gets discussed most seriously inside the hyperscalers is electricity. Not the cost of electricity. The actual physical availability of electrons - at gigawatt scale, in the places where the data centres need to be, on the schedule the model labs need them to be. The story does not have a single villain or a single number, but it has a shape, and the shape is becoming the story of the second half of the decade. ...

For most of the last decade, talking about AI hardware meant talking about Nvidia. In 2026 that has stopped being true at the inference layer. Three companies - Cerebras, Groq, and SambaNova - have built genuinely different chips around the same insight: that the workload economics of running models in production are not the same as the workload economics of training them, and that the chip architecture should follow the workload. The bet has been right enough that Nvidia has now licensed pieces of it. ...

TL;DR Document understanding is the unglamorous killer application - invoices, contracts, and scanned PDFs that were painful to extract data from are now tractable without dedicated pipelines Vision models still under-deliver on precise spatial reasoning, object counting, and subtle medical or scientific imagery - these remain jobs for specialist models Audio is the modality with the most upside: beyond transcription, it carries tone, pace, and hesitation that text loses, enabling fault detection, emotional analysis, and richer inputs The teams getting real value treat multimodal as an invisible enabling capability within a workflow, not a feature to demo - and they verify high-stakes outputs just as they would text The right question when evaluating multimodal is not “can we use this” but “what specific user problem becomes tractable that previously was not” When the first multimodal frontier models shipped, the demos were genuinely impressive. A photo of a fridge interior with the model suggesting a recipe. A handwritten napkin sketch becoming working code. A short audio clip of a meeting being transcribed, summarised, and structured. It looked, briefly, like the boundary between modalities had collapsed and we were entering a new regime in which models could reason fluidly across text, images, and sound. ...

TL;DR Prompt caching saves the computed representation of a prompt’s static prefix so subsequent requests reuse it rather than recompute it - cached tokens cost roughly 10% of normal input token prices The savings are highest when prompts have a long, identical prefix across requests - system prompts, tool definitions, and few-shot examples can make up 80-90% of total input cost The most common mistake is interpolating variables into the system prompt, which breaks caching silently; fix it by moving all static content to the top and dynamic content to the end Cache lifetimes are bounded (minutes to a few hours per provider) and any change to the prefix - including whitespace - creates a new cache miss Track your cache hit rate explicitly on every LLM dashboard; a dropping hit rate usually signals unintended prompt construction changes, and fixing it is the highest-leverage cost optimisation available If you build LLM applications for any length of time, you eventually notice that you are paying to have the model read the same instructions over and over again. The system prompt, the tool definitions, the few-shot examples, the structured output schema - all of it goes back into the model on every single request, and you pay for the input tokens every single time. For a chatbot doing one or two thousand requests a day this is annoying. For an agent doing tens of thousands of requests with long contexts, it is the dominant cost line. ...

Two years ago the way to make a model better was to train a bigger one. By the start of 2026 that recipe has stopped being the most interesting answer. The frontier has moved to a different lever - letting the model think for longer at inference time, generating intermediate reasoning, and only then producing the final answer. The category has a name now (reasoning models) and a family of products built around it. The interesting questions are no longer whether the trick works, because it clearly does, but when to reach for one, where it lands in production, and what the costs actually look like once the demo glow wears off. ...

Scott Galloway is the kind of commentator the AI conversation rarely produces: not a researcher, not a founder, not a doomer, not a booster. He is a marketing professor and a serial entrepreneur with a record of correctly reading the corporate stories of the last two decades, and he has spent the last two years pointing at the AI story with increasing concern. The headline of his pitch - that AI was not built for ordinary people and that the rich no longer need them - is provocative on purpose. The argument underneath is more careful, and worth pulling apart on its own terms. ...