Topic
44 articles tagged Writing. For the wider topic, see AI Models & News.
OpenAI cut GPT-6 API prices roughly in half and reshuffled its tier names again. What Sol and Luna cost now, and what actually changed for everyday writing.
Read article →Anthropic says Opus 5.5 writes shorter, clearer, less jargon-heavy answers. What that claim covers, what it does not, and what to check yourself.
Read article →Google shipped Gemini 3.8 Flash on 2 September 2026 with the same introductory price as 3.7 and an end date now attached. What that means for work writing.
Read article →OpenAI and Anthropic each shipped a flagship model within three days of each other. What their computer-use and pricing claims change for everyday work writing.
Read article →LinkedIn cut its own AI writer for a slop-reporting button. Substack scans posts with Pangram. What these tools actually measure, and what still publishes clean.
Read article →Dictionaries named slop the word of 2025 and readers learned the pattern. What actually triggers the label, and how to keep it off work you publish.
Read article →A Star Wars insult became the internet's shorthand for AI resentment. What that vocabulary signals about your audience, and how it should change your copy.
Read article →Most outlines are tables of contents that collapse the moment writing starts. A method that front-loads the decisions and makes the draft mechanical.
Read article →The format is not the problem. Twelve interchangeable entries written from other people's marketing pages is. How to make a list post worth publishing.
Read article →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 →A rewrite that changes a date, softens a maybe into a yes, or adds a commitment you never made is not a style improvement. It is a different message.
Read article →Courts have largely accepted that training is transformative. The live question is now what the model produces and who is liable for it, which reaches ordinary users.
Read article →The C2PA standard now covers text, and Article 50 requires machine-readable marking of synthetic content. What that actually means for the email you just rewrote.
Read article →Gallup's 2026 workplace surveys put writing and editing at the top of the use-case list. What that says about where the value is, and where the risk follows.
Read article →Article 50 transparency rules took effect on 2 August 2026. What legally requires disclosure, what does not, and how to write a policy you can actually apply.
Read article →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 →Public benchmarks measure coding and mathematics. Here is a twenty-minute test that measures whether a model will actually help with your email.
Read article →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 →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 →Small models handle some languages far better than others. How to test coverage on your own text, and which families are worth trying first.
Read article →Reasoning models deliberate before answering. For a tone change that is pure overhead. How to tell which mode you are in, and how to turn it off.
Read article →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 →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 →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 →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 →SWE-bench, MMLU-Pro, and AIME scores predict almost nothing about whether a model will rewrite your email well. What to measure instead.
Read article →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 →Not typos. The small English choices that make readers act on something different from what you intended, and how to catch them before sending.
Read article →Dictation is fast and the raw output is unsendable. The specific things speech becomes on the page, and a two-step workflow that keeps the speed.
Read article →A practical edit pass for AI-generated writing: the sentence patterns to cut, the rhythm to break, and the three checks that catch most of it in under a minute.
Read article →Repairs, deposits, and rent problems in writing. How to create a paper trail that protects you while keeping the relationship workable.
Read article →Patient portal messages are triaged in under a minute. How to structure symptoms, timeline, and your actual question so you get a useful answer.
Read article →Anger does not get refunds. Facts, a specific remedy, and a deadline do. How to write a complaint that a support agent can actually act on.
Read article →The first three lines decide whether someone keeps reading or closes the tab. What belongs at the top of a README, and what belongs much further down.
Read article →Short, specific, and one topic per message. How to raise a concern with a teacher in a way that gets a useful reply rather than a defensive one.
Read article →Group chats amplify tone and strip context. The specific habits that prevent a normal message from becoming a three-day argument.
Read article →AI assistants now summarize your page instead of sending traffic to it. How to structure writing so the machine extracts what you meant, not a confident distortion of it.
Read article →Rewriting is a constrained task, and constrained tasks favor small models. Why a 1.7B model on your laptop often produces better work email than a frontier model in the cloud.
Read article →A field guide to the punctuation, phrasing, and structural habits that make readers assume a machine wrote it, and what to do about each one.
Read article →The gap is usually register, not grammar. What actually makes work English sound off to native readers, and how to fix it without losing your own voice.
Read article →How professional writers use local AI tools not to replace their creativity, but to enhance their brainstorming and editing workflows.
Read article →Discover how moving AI assistance from a separate browser tab to an inline, on-demand tool can dramatically accelerate your writing and drafting process.
Read article →A practical method for making writing clearer while preserving the intent, nuance, and decision behind the original draft.
Read article →A simple workflow for converting messy bullets, meeting fragments, and half-written thoughts into updates people can act on.
Read article →