It is difficult to overstate how big a deal Qwen3.8 27B is as an AI model. For those new to open weights models, traditionally, open weights models have lagged behind state of the art models by 6-12 months. Using an open weights model (usually for cost reasons) meant sacrificing capabilities for cost. The team at…
AI is a pay cut for you, the human. It’s tempting to take the easy route and talk about a stagnant job market, or tech companies using AI as a cover for large scale job cuts, but that’s not what I’m talking about here. A decade ago, Katie Robbert and I worked at a PR…
This one hastily made slide explains pre-IPO jitters better than anything else when it comes to AI. DeepSeek released its v4 Flash model earlier this year, on April 24. At its release, it was widely regarded as pretty smart and stupid cheap for its size. For a lot of folks using agentic AI, its pay…
What a difference a year makes. I was putting together slides for a private workshop I’m doing today, and for fun I decided to take a snapshot of Humanity’s Last Exam results from a year ago and from today. That was eye-opening. In a year’s time, what was described as one of the most difficult…
Be careful what engine you put in your car. GPT-5.6 is unquestionably a very smart model, highly capable, up there as an Opus-class model along with Opus 4.8, Kimi K3, GLM-5.2, etc. What’s more, it’s fast, cheaper than the top models from Anthropic by an order of magnitude according to Artificial Analysis… … but it’s…
Back in 2019, Paul Roetzer asked me to speak at a brand new event, the Marketing AI Conference, MAICON hosted by the Marketing AI Institute. This was 3 years before ChatGPT, a full year before the first minimally functional GPT model even existed. How things have changed since then. The first MAICON conference was about…
One of the best things you can do in enterprise AI is set up testbeds to count tokens. Here’s why. Claude’s enterprise pricing is this: Opus 4.8: $5/million tokens in, $25/million tokens out Sonnet 4.6: $3/million tokens in, $15/million tokens out Microsoft’s pricing is based on Copilot Credits; 1 credit is one penny, and they…
Plan big, act small. With Anthropic’s latest model release, the fabulously expensive Fable 5, it’s a timely reminder that when you have a big, expensive model, you should use it to do planning, and then hand off the execution to a small, fast, cheap model that doesn’t have to think, it just has to follow…
