If you look closely at Amazon India’s robots configuration, something jumps out.

Googlebot is allowed.
Amazonbot is allowed.
Applebot-Extended is allowed.

But most major AI crawlers are blocked:

GPTBot
OAI-SearchBot
ChatGPT-User
ClaudeBot
PerplexityBot
CommonCrawl

This is not an accident.
This is not paranoia.
This is not “Amazon hates AI”.

This is Amazon telling the world:

“You can send us users.
You cannot take our intelligence.”

And that single sentence explains the future of the internet better than any keynote.

Why This Matters

Let us simplify the old internet model.

Websites created content.
Google indexed it.
Users clicked.
Websites earned through ads, subscriptions, or commerce.

Simple.

Now enter AI.

AI reads websites.
AI answers users directly.
Users stop clicking.
Websites slowly starve.

Same content.
Same effort.
Radically lower return.

So the new invisible flow becomes:

Websites → AI Models → User
Website gets nothing.

Amazon looked at this flow and said:

“No thank you.”

That is all.

No philosophy.
No ethics debate.
Only business math.

What Amazon Is Actually Doing

Amazon is not blocking everyone.

They are doing selective openness.

Search engines: Allowed
AI training and answer engines: Blocked

Translation in plain English:

“If you help people discover Amazon, welcome.”
“If you want to learn from Amazon, pay or partner.”

This distinction is important.

Google still sends traffic.
AI models mostly do not.

Amazon optimizes for incoming value, not curiosity.

Strategic Reason 1 – Protect Product Intelligence

Amazon product pages are not “descriptions”.

They are battle-tested conversion machines built from:

• Years of A/B testing
• Pricing elasticity data
• Click-through patterns
• Review sentiment clusters
• Buying behaviour at scale

This is not content.

This is retail intelligence.

Letting AI models freely ingest this means:

You are training future competitors using your own brain.

No serious platform does that.

Strategic Reason 2 – Force Paid Partnerships

Phase 1: Block.
Phase 2: Negotiate.
Phase 3: License.

We have already seen this movie.

Google paying Reddit.
OpenAI paying publishers.
Microsoft cutting content deals.

Amazon is positioning early.

Message is simple:

“If you want Amazon data in your AI, bring a contract.”

Not a crawler.

Strategic Reason 3 – Control Where Discovery Happens

Amazon wants shopping discovery inside:

Amazon Search
Amazon App
Alexa
Rufus (Amazon’s AI shopping assistant)

Not inside ChatGPT.
Not inside Perplexity.
Not inside someone else’s UI.

Whoever controls the interface controls the money.

Amazon understands interfaces better than most media CEOs understand headlines.

Global Pattern (Not Only Amazon)

This is not an Amazon-only phenomenon.

High-value platforms are quietly closing doors.

Reddit → licensing
News publishers → blocking GPTBot
StackOverflow → restricted training
X (Twitter) → tightened scraping

Pattern is clear:

Platforms with proprietary, high-signal data are locking it.

Everyone else is being harvested.

Harsh, but true.

What This Means for Publishers

Here comes the uncomfortable part.

If you are:

A news website
A niche blog
An affiliate site
A content network

And you block AI crawlers today without a licensing deal,

You are not “protecting content”.

You are committing visibility suicide.

Amazon can block because:

• They have brand demand
• They have logged-in users
• They have direct commerce revenue

Most publishers have:

Google traffic and hope.

These two things are not equal.

New Reality: Two Indexes

Stop thinking in “rankings”.

Start thinking in indexes.

Index 1 – Search Index
Google, Bing, traditional SEO

Index 2 – AI Answer Index
ChatGPT, Gemini, Claude, Perplexity, Copilot

Future visibility = Presence in both.

Amazon is choosing to dominate their own ecosystem and negotiate entry into AI.

Publishers must choose to be present everywhere.

Different games.
Different leverage.

Decision Framework – Should You Block AI Bots?

Use this simple scorecard.

AI Data Control Score (0 to 100)

Add points if you have:

Own massive proprietary dataset → +30
Large logged-in user base → +20
Strong brand demand → +20
Revenue not dependent on ads → +15
Legal & commercial team maturity → +15

If your score is above 70
You can consider blocking and negotiating.

If your score is below 70
You should stay open and optimize for AI discovery.

Most publishers score between 20 and 40.

Amazon scores in the 90s.

That is the difference.

What Smart Publishers Should Do Instead of Blocking

Blocking feels powerful.

It is mostly emotional.

Smart publishers do something harder.

They become the best source to quote.

1. Allow AI Bots

Visibility beats ideology.

Every time.

2. Add AI-Readable Signals

• Clean structured data
• Clear author bios
• About page with expertise
• Proper headings
• Source citations

If humans can skim it, AI can digest it.

3. Build “Citable Assets”

AI loves content that looks like:

• Tables
• Checklists
• Frameworks
• Step-by-step guides
• Original data

Opinion pieces get ignored.
Clear explainers get cited.

4. Track AI Traffic Separately

Create dashboards for:

ChatGPT referrals
Perplexity referrals
Gemini referrals
Copilot referrals

Treat them like early Google in 2003.

Small numbers.
Massive future value.

Prediction – What Happens Next

Over the next 24 months:

• AI crawling becomes paid by default
• Robots.txt evolves into licensing files
• Platforms sell “AI access plans”
• SEO expands into AEO (Answer Engine Optimization)

The industry will rename it.

The power shift will already have happened.

Final Thought

Amazon blocking AI bots is not rebellion.

It is clarity.

They know something most publishers are still debating:

Data is not content.
Data is leverage.

If you do not own distribution,
you must own visibility.

Right now, visibility is being decided inside AI answers.

That is the real battlefield.

And that is exactly what we focus on at rudrakasturi.com – helping publishers, platforms, and leaders win in the AI discovery layer, not just rank on Google.