We’re moving past the initial shock of the “AI arms race” and into a more complex, practical reality. The tools are getting powerfully integrated into the platforms we use every day, and the industry is starting to draw clearer lines in the sand on ethics and quality. It’s less about the theoretical and more about the “what now?”
The central question is shifting from “Can AI write?” to “How do we use it responsibly without becoming a factory for soulless content?” The answer, as I see it, lies in treating these tools not as ghostwriters, but as specialized assistants—thinking partners that help us organize, research, and see our own work more clearly, a theme I explored in my post on using AI to pressure-test my novel’s first chapter.
This week on the Substack, I took you inside my process with Opening the Door on the First Chapter. Instead of just generating prose, I used a chat with AI as a relentless editor to sharpen the worldbuilding, character, and stakes of my fantasy trilogy. The goal wasn’t to write faster, but to think better—to make sure every image and line of dialogue did real, honest work.
This craft-first approach feels more critical than ever, especially after last week’s discussion about the AI Arms Race. As the models get more powerful, our ability to direct them with intention is what will separate meaningful stories from the noise.
This week wasn’t about one single blockbuster announcement, but a series of significant moves that show how AI is settling into the fabric of our work.
Google’s NotebookLM Gets a Gemini 3 Brain: Google’s research and writing assistant, NotebookLM, has just received a major upgrade with Gemini 3, enhancing its reasoning and analysis. It has also added “Data Tables” for organizing research and expanded custom prompts.
Why writers should care: This is a move toward a true “AI thinking partner.” The ability to upload your notes, outlines, and research and have an AI help you synthesize and organize it is a massive step up from a simple chatbot.
My take: This is the kind of craft-focused tool I get excited about. It’s not about replacing the writer’s thought process; it’s about augmenting it. Use it to find patterns in your research or structure your worldbuilding bible, not to write your scenes for you.
Claude Becomes a Browser Assistant: Anthropic has released a beta Chrome extension for Claude, allowing it to navigate websites and fill out forms. This suggests a future where AI assistants are more deeply integrated into all our online tasks.
Why writers should care: Imagine an AI that can help you with the tedious parts of research—summarizing sources across multiple tabs or organizing data from web pages. It’s coming.
My take: The more integrated AI becomes, the more disciplined we have to be. It’s one thing to use a chatbot in a separate window; it’s another when the AI is woven into your browser, always ready to “help.” Define your boundaries now.
Poet Disqualified from Award for AI Use: The evolving ethical landscape was thrown into sharp relief when a poet’s collection was disqualified from the Ockham New Zealand Book Awards due to the inclusion of AI-generated content in some poems. The Trust now allows AI-generated covers but continues to prohibit AI-authored text in submissions. You can read the full story at The Spinoff.
Why writers should care: The institutions that grant awards and recognition are drawing lines. This will force a conversation about transparency and what constitutes “human-authored” work in a world of AI collaboration.
My take: This is a necessary (if painful) part of the process. Clear standards, even if they’re debated, are essential. It forces us to be more intentional about how and why we use these tools, knowing that our choices have consequences.
New Study Shows AI Misinformation Rates Doubled: A concerning NewsGuard study reported that misinformation rates in AI chatbots have doubled over the past year. While some models like Claude and Gemini performed better than others, it’s a stark reminder that AI is not a reliable source of truth.
Why writers should care: Never trust, always verify. Whether you’re researching historical facts or scientific concepts, AI-generated information requires rigorous fact-checking.
My take: Treat your AI assistant like a very confident but occasionally clueless intern. It’s great for brainstorming and summarizing, but it will state falsehoods with absolute certainty. Your credibility is on the line.
53% of Peer Reviewers Now Use AI Tools (Frontiers Whitepaper via EurekAlert)
This is a stunning statistic from the academic publishing world, but it reflects a broader trend. It’s a clear sign that AI is becoming a standard part of the professional writing and editing toolkit, forcing institutions to adapt their policies.
Chinese Hackers Used Anthropic’s Claude in Cyber Espionage (Fox Business)
While not strictly about writing, this story is a sobering reminder that these powerful tools can be used for purposes far beyond creative work, highlighting the broader societal implications of the technology we’re adopting.
This week showed that we’re in the implementation phase of the AI revolution. The tools are here, they’re being integrated, and the rules are being written (and rewritten) in real-time. The challenge for us is to stay grounded in our craft, to use these tools with intention, and to never let the machine’s voice drown out our own.
How are you navigating this new phase? Have you found a workflow that feels both effective and authentic? I’d love to hear about it in the comments.
If you’re finding these weekly roundups valuable, please consider upgrading to a paid subscription. Your support helps me continue researching and exploring the intersection of craft and technology, bringing you thoughtful insights you won’t find in the noise.
New here? Welcome to “Writing in the Age of AI: A Craft First Approach.” This newsletter is dedicated to helping you maintain an authentic voice and elevate your craft in a world where machines can write too. No hype, no panic – just a focused exploration of what it means to write well.
No posts
Comments
Nothing yet. Say the first thing.
Sign in to join the conversation.