5 min read
Why Two Leaderboards Rank The Same Model Differently
One board has a model first, another has it fifth, and both are honest. The four reasons scores diverge, and which board to trust for writing work.
Read article →AI Models & News
Model releases arrive faster than anyone can evaluate them, and most announcements describe capabilities that matter enormously for building software agents and very little for writing a difficult email. These articles separate the two: what each release actually changes, which specifications are marketing, and how to keep a working setup without chasing every launch.
29 articles
Narrow it down
6 min read
Why AI Model Version Numbers Tell You Nothing
GPT-5.6, Gemini 3.6, Claude Sonnet 5, GLM-5.2. The numbers are marketing, not measurement. How to work out whether a release actually matters to you.
Read article →6 min read
The Em Dash Is Not Proof Of AI Writing
The internet decided one punctuation mark exposes machine writing. The measurements say otherwise, and the real tells are harder to spot and harder to fake.
Read article →6 min read
AI News, Early August 2026: The Five Things That Actually Matter for Work Writing
A month of model launches, price changes, and a regulatory deadline. What genuinely changes if your job involves writing emails and documents, and what is noise.
Read article →7 min read
AI Provider Outages Are Routine Now: Designing a Workflow That Survives Them
A major model provider logged a multi-hour outage in early August 2026. What that means if your writing depends on a hosted model, and what a real fallback looks like.
Read article →6 min read
Claude Sonnet 5 Promotional Pricing Ends 31 August 2026: Budgeting for Volatile AI Costs
A mid-tier model gets 50 percent more expensive overnight when a promotion expires. What that says about planning AI spend, and how to build a setup that survives it.
Read article →6 min read
Meta Went Closed: What Muse Spark Means for Open Weights
Meta shipped Muse Spark 1.2 in August 2026 behind an API and has not released a new open Llama in over a year. Who carries open weights now, and what it changes for you.
Read article →7 min read
OpenAI Cut GPT-5.6 Prices in August 2026: What Cheap Inference Changes
An 80 percent cut on the cheapest tier and 20 percent on the middle one. What falling token prices actually change for work writing, and what they do not.
Read article →6 min read
Qwen3.8-Max: What a 2.4 Trillion Parameter Model Changes for Your Writing
Alibaba's largest model yet lands in August 2026 with open weights promised. Here is what the specifications mean and why your rewrite still runs on 1.7 billion parameters.
Read article →6 min read
ChatGPT Work and Agents That Finish the Job: What It Changes for Writing
OpenAI shipped an enterprise agent in July 2026 that runs for hours and produces finished documents. What that changes for professional writing, and what it does not.
Read article →6 min read
Google's Cheap Gemini Tier Got Serious: What Flash Models Mean for Writing
Three new Gemini models shipped in July 2026, all in the cheap and fast tier, none a new Pro. Why the workhorse tier is where the useful work now happens.
Read article →6 min read
Token Efficiency Became the Benchmark Labs Advertise in 2026
Providers now market doing more with fewer tokens rather than higher scores. Why that shift happened, and why it is better news for writing than for anything else.
Read article →7 min read
Agentic AI in 2026: 79 Percent Adopted, 11 Percent in Production
Enterprises say they have adopted AI agents. Very few are running them in production. What the gap tells you about where AI actually delivers value today.
Read article →7 min read
AI Browser Agents and Your Clipboard: The New Exposure Surface
Agents that read your screen, your tabs, and your clipboard transmit far more than you intend. What that means for confidential work, and how to draw a boundary you can hold.
Read article →7 min read
AI Writing Detection in 2026: Where Things Actually Stand
Detectors are still unreliable, watermarking is partial, and the professional norms have shifted. A current picture of what can and cannot be proven about AI-assisted text.
Read article →6 min read
Keeping Facts Accurate When the Model Underneath You Changes
The most dangerous rewrite is the fluent one where a figure moved. How to build a checking habit that survives model upgrades, deprecations, and silent version swaps.
Read article →6 min read
Long Context Quality Degradation: Why Big Windows Lie a Little
Models accept a million tokens and attend well to far fewer. Where quality drops, why the middle of a long input is the danger zone, and how to work around it.
Read article →6 min read
Prompt Caching and Batch Pricing, Explained
Two pricing mechanisms that cut cloud AI costs substantially, when each applies, and why neither helps the interactive rewrite you are waiting on.
Read article →7 min read
Sovereign AI and Where Your Text Is Actually Processed
Data residency, jurisdictional control, and operational access are three different things. What sovereignty claims mean, and the configuration that settles all three.
Read article →6 min read
AI Models Are Becoming Interchangeable. Plan Accordingly.
Prices fell roughly 80 percent in a year and open models sit within a few points of proprietary ones. When the model is a commodity, the durable decisions are about everything else.
Read article →6 min read
Four Models in Two Months: What the 2026 Release Cadence Means for You
Anthropic shipped four Claude 5 models in under two months. AI has moved from blockbuster launches to continuous iteration, and that changes how you should build workflows on top of it.
Read article →6 min read
Chinese AI Models in 2026: What Professionals Should Actually Know
DeepSeek, Qwen, Kimi, GLM, MiniMax, and Hunyuan now define the open-weights frontier and undercut Western pricing dramatically. The capability story, and the procurement questions it raises.
Read article →6 min read
Claude Opus 5 for Professional Writing: What Actually Changed
Anthropic shipped Claude Opus 5 in July 2026 with a million-token context, adaptive thinking, and effort levels. Which of those features matter for work writing, and which are for other jobs.
Read article →6 min read
Do Frontier Models Write Better Emails? A Careful Answer
The best model in the world and a 1.7B model on your laptop, given the same email. What actually differs, what does not, and why the expensive one sometimes loses.
Read article →6 min read
Do You Need a Million-Token Context Window?
Million-token windows are now standard on frontier models. What that capacity is actually for, why retrieval quality degrades long before the limit, and what professional writing really needs.
Read article →6 min read
GPT-5.6 Explained: Luna, Terra, and Sol for Everyday Work
OpenAI's GPT-5.6 ships as three tiers with different capability and cost. Which one you actually want for professional writing, and why the answer is usually not the flagship.
Read article →7 min read
How to Keep Up With AI Model Releases Without Losing Your Week
Four frontier models in two months and a new open-weights release most weeks. A filtering system for staying current on AI without turning it into a second job.
Read article →6 min read
OpenAI vs Anthropic for Workplace Writing in 2026
Both labs ship excellent models. For rewriting work messages the differences that matter are not the ones the benchmarks measure. A practical comparison, including where neither is the answer.
Read article →6 min read
What to Do When Your AI Model Is Deprecated
Every hosted model you build a habit on will eventually be retired. How to notice early, migrate without breaking your prompts, and keep one option that cannot be taken away.
Read article →