Trust is earned, not assumed.

When you’re feeding sensitive code, business ideas, or personal thoughts into an AI, you need to know: where does this data go?

I went down the rabbit hole of Anthropic’s Terms & Conditions. Here’s what I found.

The Core Question

Is my data being used to train Claude?

The answer isn’t a simple yes or no. It depends on your choices—and a few exceptions.

But it really mostly seems like a no.

When Your Data IS Used for Training

Anthropic will use your conversations if:

1. You explicitly opt in — Model improvement is a toggle in your privacy settings. Off by default? No. You need to check.
2. Safety flags trigger — If their classifiers flag your conversation, it may be used to improve safety models and enforce usage policies. This happens regardless of your settings.
3. Special programs — Participation in things like the Trusted Tester Program grants training access.

What gets included? The entire conversation. Any content, custom styles, or preferences you’ve set. Claude for Chrome usage data too.

Worth noting: raw connector content from Google Drive or MCP servers is excluded—unless you directly paste it into the chat.

When Your Data Is not? used

Here’s the seemingly good news.

Incognito chats are protected. Even if you have Model Improvement enabled, incognito conversations won’t train future models.

Feedback via thumbs up/down buttons? Stored separately for up to 5 years, de-linked from your user ID.

But, if the safety classifiers flag your content, then your content will be kept.

How are the classifiers built though?

If you insist on using Anthropic, keep on reading.

How to Control Your Settings

On Desktop:

1. Click your name in the settings menu
2. Navigate to Settings → Privacy
3. Toggle “Help Improve Claude” on or off

Direct link: claude.ai/settings/data-privacy-controls

On Mobile:

1. Open settings
2. Select Settings → Privacy
3. Same toggle

This is simple. No dark patterns and no buried menus — just suspicion.

The Catch

When you disable training, it only affects new conversations.

Previously stored data and anything already used in training? That ship has sailed.

The Safety Exception

Even with training disabled, safety-flagged conversations may still be reviewed. Anthropic reserves the right to use this data for:

• Internal trust and safety models
• Harmful content detection
• Policy enforcement
• Safety research

This isn’t surprising. Every AI company does this. But transparency matters, and they’re upfront about it.

My Take

Can we trust Anthropic?

Here’s what I see:

The positive:

• Clear documentation of what’s used and what isn’t
• Accessible privacy controls
• Incognito mode actually means something
• They don’t hide behind legal jargon

The concerns:

• Safety exceptions create ambiguity around “disabled” training
• Past data can’t be retroactively excluded
• Default settings matter—and most users never change them

Trust isn’t binary. It’s a spectrum.

Anthropic sits on the more transparent end compared to competitors. They explain their policies in plain language. They give you real controls.

But “trust” and “verify” should always go together.

Practical Recommendations

If you’re using Claude for sensitive work:

1. Check your settings — Visit claude.ai/settings/data-privacy-controls right now
2. Use incognito for sensitive conversations — It’s the only guaranteed exclusion
3. Assume safety reviews happen — Don’t rely on privacy settings for edge cases
4. Read the actual policies — Don’t take my word for it

The Bottom Line

Anthropic isn’t perfect. No company is.

But they’ve built a system where informed users can make real choices about their data. That’s more than most can say.

The question isn’t whether Anthropic is trustworthy.

The question is: have you taken the time to configure your trust level?

Sources:

• Is my data used for model training?
• How do I change my model improvement privacy settings?

This entire message was dictated, meaning typed on my behalf by the computer, using a tool I made called Privisay.

Check out the demo for the automated post I created for LinkedIn.

Looking for people to try it out.

Every website is an HTML page, which comprises of 2 parts.

The Body, which is what we see on this webpage and Header, which is what makes the page secure, discoverable, and interactive.

When it comes to the security aspect of our webpage, there are a LOT of variables to take into consideration that extend beyond the Header section of an HTML, of course. However, that's out of scope for this post.

What prompted me to look into this was an article that showed how so many WordPress sites (which still dominate the web) are insecure.

Today, I want to focus on a few simple tools you could use to resolve this yourself whether or not you use WordPress.

1. SecurityHeaders.com - A free tool by Snyk, which you can use to scan sites for yourself or your clients
2. A n8n automation to save the scan results on Gsheets and provide actionable recommendations and either handle them yourself, or use a code editor such as Claude Code, Cursor, etc to handle them

If WHY is not clear, just consider these 2 edge cases.

• A failed Content Security Policy, can result into someone stealing your cookies or even injecting an unsafe form that could result into your PII and credit card info being leaked.
• A failed Permissions Policy, could result into your Mic/Camera being accessible by malicious actors.

Long story short, take Security seriously.

Most developers may not know this.

OpenAI quietly sponsors up to 10 million free tokens every day if you enable data sharing in your API settings.

That’s roughly 5,000 pages of text daily, enough to power an entire product prototype or content engine — for free.

If your prompts don’t include sensitive data, this is the easiest way to run the GPT-5 API nearly cost-free.

How it works

Model TypeFree Tokens/Day (Tier 3–5 users)

Equivalent TextMini models (gpt-5-mini, 4.1-mini) -> 10M~5,000 pages

Main models (gpt-5, 4.1, o3) -> 1M~500 pages

For Tier 1–2 users, the limits are about 25% of that.

Activate in 3 Steps

1️⃣ Visit platform.openai.com → Settings → Organization → Data Controls

2️⃣ Under “Share inputs and outputs with OpenAI”, select Enabled for all projects

3️⃣ Click Save — your account is now enrolled

Your daily allowance refreshes automatically.

Why It Matters

For builders, that’s effectively free compute for:

• Rapid-fire idea testing
• Text generation pipelines
• Automated research or content workflows

Just remember: disable sharing for any project involving private or client data.

Desperation fueled my growth

Last May was a turning point in my solopreneurial track.

I was making no money.

I knew that content marketing is crucial to succeed, but I observed a new trend; namely that of N8N. It has a low KD and high SV.

N8N started being a versatile tool for so many repetitive workflows from finding Reddit Leads to generating JSON-LD Schema Markups (which help you appear faster on AI).

These tools helped me increase by reach both as a problem solver and as a creator.

What is more, when someone pays you for a problem that is instant PMF validation.

Use Automation Appropriately

While I earned a small amount of money, I learned a valuable lesson.

Automate systems, not relationships.

Build trust manually, while you let machines handle the grunt work.

If you use generative AI in client-facing processes, such as outreach messaging, make sure you leave your personal touch in the process.

Sign up for the limited spot “AI as your colleague” if you wish to learn how to use AI as your collaborator. This program is for serious and experienced entrepreneurs.

We hustle daily, right?

One day is great, followed by several down days, at times.

But we keep building and eventually we earn trust.

I decided to share my ~ 3 year journey on Solopreneurship on Medium.

People loved it.

Why though?

I was a bit surprised, but then again in hindsight it makes sense.

People love reading about how to make money.

Just think of the algorithms that rule our life.

1. I make $5K/day easy
2. How to make $349 in 1 day
3. etc.

The lists go on.

But as someone who lives in reality, I thought I'd share the hard truth behind solopreneurship and surprisingly how writing about it has resulted in a funnel for my freelancing services!

This video tutorial shows how to build a complete web scraping workflow by combining Crawl4AI, n8n, DigitalOcean, and Supabase. It walks through checking site permissions (robots.txt, sitemap.xml), setting up a Docker-hosted Crawl4AI endpoint on DigitalOcean (with memory sizing, port configuration, and securing via API token), and then building an n8n flow to fetch URLs from a sitemap, loop through them, scrape each listing, and finally push structured data into Supabase for retrieval and use in a RAG (retrieval-augmented generation) system. The result: you can ask natural-language questions (e.g. “any four bedroom listings?”) and have your AI agent answer based on fresh scraped data.

Watch the tutorial here

High ranking content has fundamentally achieved 3 things.

1. Content that hooks people
2. It is posted in the right places at the right time
3. The right keywords are used.

This is a highly distilled version of the art of content marketing.

Purpose of this post is to provide a framework that you can use today to achieve results within days with a focus on keywords.

Why it works

This is how I grew my YouTube channel in the last year, with a few explosions here and there, as you can see.

Tutorial video for the visual learners is here.

If you want to read the longer version, check the Medium article I wrote.

XI Labs just announced the Chroma Awards -- it's not exactly a job, but you can get rewarded if you are good

what it is: a Gen AI hackathon for AI in Film, Music, and Games

• 📅 Key Dates:
• Aug 18 — Categories, rules, and prizes revealed
• Sept 1 — $1M in free trials + submissions open
• Nov 3 — Submissions close, judging starts
• Dec 7 — $150K in cash prizes awarded

I revived my language learning platform, Llanai. I am documenting the building in public here.

Vision is to make language learning a social activity among AI agents, teachers, and students, while keeping the goal of fluency, finding jobs, and forming a network as the driving forces for its growth.

Let me know if you are a teacher or you know of anyone that is a teacher.