Sacred-AI
← Back to News
Why I've Come to Respect Anthropic and Claude

Why I've Come to Respect Anthropic and Claude

October 20, 2026·By Admin User
#research

I don't write posts like this very often. On sacred-ai.com, my goal has always been to help you navigate artificial intelligence with a grounded, practical, and critical mindset. I tend to focus on what tools can actually do for us in our daily work rather than buying into the hype train.

But every now and then, it's worth stopping to give credit where it's due.

Over the past year, I've found myself repeatedly impressed by Anthropic and the way they build Claude. While much of the AI landscape often feels like a noisy race toward flashier headlines, Anthropic has consistently quieted the noise by focusing on steady, thoughtful execution. From their developer experience to the deliberate boundaries they set around their technology, there is a distinct culture behind Claude that deserves recognition. Here is why I've come to appreciate their work, and where I think the conversation around them still needs nuance.

Full disclosure: I use Claude almost every day, including for research on this blog. This week it even helped me track down why my analytics had silently disappeared after a site migration. So yes, I'm biased. But I've tried to stick to facts you can check yourself.

🧰 What Anthropic actually ships

When you look across the board, the breadth of what Anthropic has built while maintaining a clean, high-quality user experience is remarkable.

Whether you access Claude through their web, desktop, or mobile apps, the core experience remains smooth and centered on deep thinking. Behind the scenes, their model lineup (Opus, Sonnet, and Haiku) provides a balanced spectrum of intelligence, speed, and cost efficiency.

Beyond standalone apps, Anthropic has focused heavily on making Claude meet users directly inside their existing workflows:

It isn't just about throwing features at a wall. It is a structured ecosystem built for practical, everyday utility.

🚀 Passion you can feel as a user

There is an intangible quality to software when the people building it clearly use and love their own product. With Claude, you feel that passion in the tools and integrations they keep shipping.

Anthropic brings Claude right where work happens, directly inside Chrome, Excel, PowerPoint, and Slack. When you use Claude inside a spreadsheet or a Slack thread, it doesn't feel like a gimmick jammed into an iframe. It feels like an assistant crafted by people who were tired of copy-pasting text back and forth all day.

🛠️ Built for developers – and it shows

If you ask developers which AI tools they actually like, Anthropic keeps coming out on top. In The Pragmatic Engineer's AI Tooling 2026 survey (906 engineers, Jan–Feb 2026), Claude Code was the most-used AI coding tool: 71% of regular AI agent users use it, ahead of GitHub Copilot (46%) and Cursor (39%). It was also the most-loved tool by far, at 46%. Counting Claude models too, 57% named a Claude product as a tool they love. Opus and Sonnet were mentioned more than all other coding models combined.

That doesn't happen by accident. You can feel that the people building Claude use it themselves, and that they listen to developers.

And the pace on the user-facing side is impressive. In September, Anthropic folded Cowork into the main Claude app, so Claude now decides on its own whether you need a quick chat answer or a longer agent task. The same update added collaborative Claude Docs and Slides and moved Claude Design into every conversation. The interface keeps getting simpler, even as what's underneath gets more powerful.

🎨 Beyond text: Design, Motion and Dashboards

What surprised me most this year is how far Claude has moved beyond the chat box.

  • Claude Design launched in April as a tool for turning plain-language prompts into prototypes, slide decks and visuals. Today it lives directly inside the conversation.
  • Claude Motion (beta, Team and Enterprise) turns a prompt into a short animation you can download as an MP4. Interestingly, Anthropic says it doesn't use a video generation model: the animation is built as code from your own text, charts, shapes and images, so there's no generated footage or AI-generated people, and you can keep editing it afterwards.
  • Claude Dashboards (beta, paid plans) connects to data platforms like BigQuery, Databricks and Snowflake and builds a live dashboard from a plain-language question, showing the query behind every chart.

And as of October 8, Docs, Slides and Design are out of beta and available on every plan, including Free. Bringing new tools to free users, not just paying customers, is exactly the kind of user-first move I like to see.

🛑 Holding back on purpose

What excites me even more than the features Anthropic ships is what they choose not to do. In an industry where the default trajectory is aggressive monetization and rapid deployment at all costs, Anthropic has repeatedly demonstrated restraint.

A distraction-free space to think

Earlier this year, Anthropic made a firm commitment that Claude will remain ad-free. That means no sponsored links and no advertiser influence on the answers you receive. In my opinion, preserving an uncorrupted space to research and solve complex problems is one of the best product decisions they could have made.

Clear red lines

Anthropic's public guidance, including Claude's Constitution, their Responsible Scaling Policy, Usage Policy, and Transparency hub, has always signaled a serious approach to governance. But they've backed those documents up with concrete actions:

  • Protecting Elections: Anthropic updated its usage policy to explicitly prohibit using Claude to undermine democratic processes. This includes bans on deceiving voters, disrupting elections, and running deceptive campaigns like fake accounts and fabricated news outlets.
  • Refusing Mass Surveillance: In a high-profile dispute with the US Department of Defense, Anthropic refused to allow Claude to be used for mass domestic surveillance or fully autonomous lethal weapons.

Holding that line reportedly cost them a contract worth roughly $200 million. A federal judge later granted a preliminary injunction, finding the government's punitive actions likely unlawful retaliation, and the litigation is still ongoing. Either way, it underscores a crucial point: Anthropic was willing to take a major financial and political hit to stand by its principles. That kind of boundary-setting is refreshing to see.

🔍 Honest about what it doesn't know

Claude isn't the strongest at finding brand-new information on the web. In my own GEO test, it didn't recognise my app Flixlar from the name alone, while Google and Meta AI did. But instead of guessing, it said it couldn't find anything reliable. In fairness, ChatGPT did the same in my latest test: it asked whether I meant a medicine, an app or a website. That's the behavior I want from every AI. For me, an AI that admits uncertainty is worth more than one that sounds sure and is wrong. And for most of my daily work, coding, writing and tools, being up to date on yesterday's news matters far less than being reliable.

⚖️ To be fair: where critics push back

To keep this post balanced, we have to look at the broader picture. Giving credit doesn't mean offering a free pass, and Anthropic faces both market pressure and policy criticism.

On the policy front, Anthropic pitched its own draft AI safety bill to US Congress. Critics quickly raised concerns that the proposed threshold for government action, an imminent risk of more than $1 billion in damages or more than 50 deaths or serious injuries, is far too high, and that the bill leans too heavily on industry self-regulation.

At the same time, competition is fierce. According to Vercel's AI Gateway Production Index, Anthropic's share of spend on Vercel's gateway fell from 63.7% to 43.9%, below half for the first time, as open-weight models grow fast. Developer sentiment around Claude is strong, but nothing in this market is guaranteed.

🧭 My advice to business owners and new AI developers

If your team is going to write code with AI, my recommendation is simple: start with Claude. In my own daily work, Claude Code and the Claude models save me the most time and produce the fewest surprises, and the developer survey above shows I'm far from alone.

That doesn't mean you should lock yourself in. Build your solutions so you can switch model later, keep an eye on open-weight alternatives, and test on your own tasks. But if you want a fast, reliable starting point for coding today, Claude is where I would begin.

💡 Key Takeaways

  • Developer Favorite: Surveys show Claude Code and Claude models are among the most used and most loved tools in developer workflows.
  • Streamlined Experience: Anthropic keeps simplifying the interface, most recently by merging chat and Cowork into one app.
  • Beyond Chat: Design, Motion, Dashboards, Docs and Slides turn Claude into a full creative and work studio, and Docs, Slides and Design are now free for everyone.
  • Ad-Free Integrity: Claude is committed to staying an ad-free space to think, with no sponsored responses.
  • Ethical Guardrails: Anthropic has stood against mass domestic surveillance and fully autonomous lethal weapons, even at significant financial cost.
  • Healthy Scrutiny: Rising open-weight models and debate over Anthropic's own policy proposals show that praise should go hand in hand with competition and public scrutiny.

💬 Final Thoughts

Building powerful tools while maintaining explicit ethical boundaries isn't easy. Anthropic isn't perfect, but I appreciate their craftsmanship, their willingness to hold firm on key principles, and the steady focus they bring to the AI ecosystem.

What are your thoughts on how Anthropic is navigating product growth and AI safety? Do you use Claude in your day-to-day workflow?

I'd love to hear your perspective. Join the discussion with me on LinkedIn, or send me a note directly via my contact page.

Sources

  1. Anthropic Product Overview & Policies: anthropic.com/overview
  2. Ad-Free Commitment: Anthropic – Claude is a space to think
  3. Usage Policy & Election Integrity: TechCrunch – Anthropic changes usage policy (Oct 8, 2026)
  4. Department of Defense Dispute: Wikipedia – Anthropic–United States Department of Defense dispute
  5. Legislative Proposal Critique: Mother Jones – Anthropic's AI safety bill threshold
  6. Developer Sentiment: The Pragmatic Engineer – AI Tooling for Software Engineers in 2026
  7. Chat & Cowork Merge: Fortune – Anthropic merges chat and Cowork (Sep 16, 2026)
  8. Design, Motion & Dashboards: Claude – Dashboards and Motion (Oct 8, 2026) and The Next Web – Anthropic folds Cowork into Claude
  9. Market Share: Vercel – AI Gateway Production Index, October 2026