- Horizon AI
- Posts
- Write and edit Word docs smarter with Claude 🚀
Write and edit Word docs smarter with Claude 🚀
Microsoft Is Restricting Employees from Using Claude Fable 5 😱
Gina Acosta
June 11, 2026
Welcome to another edition of Horizon AI,
In today's issue, we share how to get started with Claude for Word to collaborate with Claude directly where you work and create, edit, and improve documents faster with AI.
Let’s jump into it!
Read Time: 4.5 min
Here's what's new today in the Horizon AI
Microsoft Restricts Claude Fable Internally Over Data Retention Concerns
Anthropic CEO Calls for Government Power to Block Dangerous AI Systems
AI Tutorial: How to use Claude for Word to write and edit docs faster
AI Tools to check out
AI Findings/Resources
The Latest in AI and Tech 💡
AI News
MICROSOFT
Microsoft Restricts Claude Fable Internally Over Data Retention Concerns
Microsoft is reportedly limiting employee access to Anthropic's newly released Claude Fable 5 model as its legal teams review changes to the model's data retention policies.
Details:
While Microsoft quickly rolled out Claude Fable 5 to GitHub Copilot and Foundry customers, the model remains unavailable on the internal versions of GitHub Copilot that Microsoft employees use.
The concern centers on Anthropic's new data retention requirements, which differ from the Zero Data Retention policies used by previous Claude models.
To support the new safety systems of its first Mythos-class AI model, Anthropic retains prompts and outputs submitted to Claude Fable 5, holding the data for up to 30 days before deleting it.
In some cases, content flagged for potential policy violations may be retained for up to two years.
Microsoft's legal teams are evaluating how these retention policies could affect the handling of customer information, confidential data, and internal company content, and it is not yet clear whether the model will ultimately be cleared for internal use.
TOGETHER WITH HEAR.COM
Brain-Boosting Hearing Innovation
When your hearing struggles, your brain works overtime to fill in the gaps, which can affect memory, focus, and energy over time. Horizon IX from hear.com lightens that load. This tiny, virtually invisible device uses advanced German dual‑processing technology to separate speech from background noise, so conversations feel effortless again. Built for modern life and all‑day comfort, it’s available with a 45‑day no‑risk trial so you can hear the difference for yourself.
AI REGULATION
Anthropic CEO Calls for Government Power to Block Dangerous AI Systems
Anthropic CEO Dario Amodei has proposed a new framework for AI regulation, arguing that governments should have the authority to stop the deployment of advanced AI systems that fail to meet strict safety standards.
Details:
Amodei says AI development is advancing much faster than public policy, creating an urgent need for stronger oversight of increasingly capable systems.
He proposes mandatory safety testing for frontier AI models, particularly in areas such as cybersecurity, biological risks, autonomous behavior, and advanced research capabilities.
Under his framework, AI systems would undergo evaluations similar to those required for industries such as aviation, pharmaceuticals, and automotive safety.
Models that fail to meet established safety thresholds could be blocked from release or withdrawn from deployment if they are deemed a threat to public safety.
Amodei argues that current transparency-focused regulations are not sufficient and that more binding and enforceable AI rules may be necessary as capabilities continue to improve.
Anthropic is also calling for policies to address the broader economic and societal impact of AI (including wage insurance, tax incentives for retaining workers, and expanded social safety programs), as well as faster regulatory pathways for beneficial applications like AI-assisted medical discoveries, and restrictions on fully autonomous weapons.
AI Tutorial
How to use Claude for Word to write and edit docs faster
Open your document and click the Claude logo in the upper right corner to open the Claude sidebar (if it doesn't open automatically after installation).
You'll be asked to log in to your Claude account.
You can ask Claude to create content from scratch, request rewrites or summaries of whole documents, or highlight specific parts to have Claude make targeted edits.
Examples:
"Rewrite this in a more formal tone"
"Make this paragraph shorter and punchier"
"Simplify this for a non-technical audience"
It can also automatically format the document, add headings, bullet points, and more.
You can follow up for further changes until it's exactly what you want.
Note: Claude for Word is available on all paid Claude plans.
AI Tools to check out
✍ Jenni: An AI research and academic writing assistant that helps students, researchers, and academics create essays, papers, and citations with ease.
💼 SelectPrism: Cut time-to-hire by 35% with AI that screens, schedules, and conducts first-round interviews.
👉 ASI:One: A customizable personal AI designed to learn, socialize, and take action with everyday tasks.
📘 1Scholar: All-in-one AI academic toolbox.
🚀 Replifine: Turn screenshots into pixel-perfect React, Next.js, and Vue code 4x faster.
TOGETHER WITH GREENFIELD ROBOTICS
What Replaces Roundup?
The next agricultural transition may not be bigger tractors. It may be autonomous robots replacing herbicides entirely. Greenfield Robotics is building commercial systems designed for that future.
Greenfield Robotics is Testing The Waters under tier 2 of Regulation A. No money or other consideration is being solicited, and if sent in response will not be accepted. No offer to buy the securities can be accepted and no part of the purchase price can be received until the offering statement filed by the company with the SEC has been qualified by the SEC. Any such offer may be withdrawn or revoked, without obligation or commitment of any kind, at any time before notice of acceptance given after the date of qualification. An indication of interest involves no obligation or commitment of any kind. “Reserving” shares is simply an indication of interest. There is no binding commitment for investors that reserve shares in this manner to ultimately invest and purchase the shares reserved of the company, or to purchase any shares of the company whatsoever.
AI Findings/Resources
👀 Microsoft AI CEO Mustafa Suleyman criticized Anthropic for speculating about Claude's consciousness, calling it "really, really dangerous"
🤔 People are using AI to communicate without disclosing it. Is this morally wrong?
The latest in AI and Tech
Anthropic has revised a controversial policy tied to its new Claude Fable 5 model after significant backlash from the AI research community and even former Anthropic employees. The company had originally planned to quietly reduce the model's performance when it detected users were using it for advanced AI development, a move critics described as hidden interference with legitimate research.
Following the backlash, Anthropic said it will make those safeguards visible. Instead of secretly degrading performance, the company will now notify users when requests are refused or redirected to a less capable model.
Google DeepMind has released DiffusionGemma, a new open-source AI model that takes a different approach to text generation than traditional language models. Instead of generating text one token at a time, DiffusionGemma generates entire blocks of text simultaneously, delivering up to 4x faster text generation on GPUs.
Elon Musk has shared a new look at the orbital AI data centers that SpaceX plans to deploy in space, offering the most detailed preview yet of the company's ambitious project. The proposed satellite would stand about 20 meters tall with a 70-meter wingspan, making it the largest satellite SpaceX has launched to date.
OpenAI CEO Sam Altman reportedly told employees that he expects the company to go public within the next year, according to a report from The Information. The OpenAI leader also signaled that a new AI model is nearing release.
China is preparing a massive 2 trillion yuan ($295 billion) investment plan to expand AI infrastructure nationwide, aiming to strengthen domestic AI capabilities and reduce reliance on U.S. technology. The proposal would fund a network of interconnected data centers across the country, operated primarily by state-owned telecom companies and powered largely by Chinese technology, with local suppliers expected to provide at least 80% of key components such as AI chips.
That’s a wrap!
We'd love to hear your thoughts on today's email!Your feedback helps us improve our content |
Not subscribed yet? Sign up here and send it to a colleague or friend!
See you in our next edition!
Gina 👩🏻💻