Mike Caprio
Verified account This account is verified. Learn more@michaelcaprio
Media, Technology & Video | Chief Growth Officer @JWXofficial | Former @Sizmek @ZetaGlobal @Disney @CBS | Father. Sports Junkie. Business Builder.
2.1K Following 1K Followers 2K Posts
Latest posts 5
NYC is going to be insane next week and kicks off an unbelievable sports stretch. World Cup. Yanks/Dodgers. US Open at Shinnecock.
@nyknicks come home to the Garden up 2-0. @NBA tried their best Vince McMahon script last night. Didn’t happen.
Let’s finish it 💪
@nyknicks come home to the Garden up 2-0. @NBA tried their best Vince McMahon script last night. Didn’t happen.
Let’s finish it 💪
Cloudflare processes 7–11 trillion internet requests per day. 20% of all global web traffic. Every request logged, classified, flagged in real time.
Not surveys. Not models. The actual internet.
We pulled their US data from the last 90 days.
👉 45% of all traffic is automated
👉 71% of actual web page requests are bots
👉 416 billion AI scraping requests blocked since July 2025 — just the ones they stopped
The open internet is realigning from human-to-machine to machine-to-machine. Faster than most thought.
Agent-to-agent is the Silicon Valley pipe dream. But it's already happening indirectly.
✔️ Mom asks ChatGPT what stroller to buy
✔️ Value shopper asks Perplexity for coupon codes
✔️ Traveler asks Claude to find the cheapest route to Fiji
The LLM crawls the open web to answer. Learns. Builds context. Never sends the click back.
Anthropic: 500,000 crawler requests per referral. OpenAI peaked at 3,700:1.
The content gets extracted. The visit never happens. The publisher gets nothing.
If publishers thought adblocking was bad — this can be worse.
The entire open web was built on one assumption: humans engage, humans see ads, humans buy things.
That assumption is eroding. Now.
The future may quietly be here already. Are you prepared for it?
Not surveys. Not models. The actual internet.
We pulled their US data from the last 90 days.
👉 45% of all traffic is automated
👉 71% of actual web page requests are bots
👉 416 billion AI scraping requests blocked since July 2025 — just the ones they stopped
The open internet is realigning from human-to-machine to machine-to-machine. Faster than most thought.
Agent-to-agent is the Silicon Valley pipe dream. But it's already happening indirectly.
✔️ Mom asks ChatGPT what stroller to buy
✔️ Value shopper asks Perplexity for coupon codes
✔️ Traveler asks Claude to find the cheapest route to Fiji
The LLM crawls the open web to answer. Learns. Builds context. Never sends the click back.
Anthropic: 500,000 crawler requests per referral. OpenAI peaked at 3,700:1.
The content gets extracted. The visit never happens. The publisher gets nothing.
If publishers thought adblocking was bad — this can be worse.
The entire open web was built on one assumption: humans engage, humans see ads, humans buy things.
That assumption is eroding. Now.
The future may quietly be here already. Are you prepared for it?
@NJTRANSIT, your new app sucks. If you are interested in how your users actually use your app and will take that feedback to correct the UX, please dm me.