In partnership with
Google Is Crawling Harder And Sending Nobody
Reddit reported earnings on Thursday.
Revenue was up 53%.
The stock fell 18% anyway.
Here’s what happened.
Google search referrals to Reddit dropped by more than 1 million a week in October.
CEO Steve Huffman put it on Google’s AI Mode and AI Overviews, which answer the question on the results page so nobody clicks through.
His line to analysts: “The reality is, the AI mode does not drive traffic, not for us and not to anybody.”
Think about that.
The most-searched site on Google just said the pipe is shut.
US daily users fell 2%, Reddit’s first decline since it went public in 2024.
If October’s trend holds, Huffman warned, daily users could drop by a low double-digit percentage next quarter.
Here’s the detail that should bother you.
Google is still crawling Reddit as hard as ever.
Translation: the content gets taken, the answer gets shown, the visit never happens.
Same day, Google admitted on its own call that AI Mode ads earn less than regular search.
So the clicks aren’t moving to Google’s wallet either.
They’re vanishing. 🙄
Reddit’s answer is the oldest one in the book.
Make people log in.
Most of the site is going behind a sign-in wall, which stops scrapers and turns strangers into members in one move.
Logged-in users are already 80% of Reddit’s money.
Let me explain why this is your problem, not Reddit’s.
This week Microsoft cut a quarter of its staff to fund AI.
Google cut a quarter of the internet’s referrals to fund the same thing.
Every business that built its lead flow on “people search, people click” just got told by the biggest forum on earth that the second half of that sentence is over.
The fix is the same one Reddit picked.
Stop renting attention from Google and start owning it.
Speaking of owning things...
Reddit’s whole plan is to stop depending on a platform it doesn’t control.
Same rule applies to the AI tools you use every day.
A prompt you typed once is rented.
A workflow you built and saved is owned.
I put together a free 53-page guide to using Claude, Cowork and Claude Code like a pro: build reusable workflows, set safe boundaries, and get results you can trust.
Google can take your clicks.
Nobody can take a workflow you built yourself. 😏
Did You Know?
In 2024 Google agreed to pay Reddit about $60 million a year to train its AI on Reddit’s posts.
Two years later that AI answers the question and keeps the visitor.
Reddit got paid to build the thing that stopped sending it traffic. 😂
The ice cream shop that makes money when it's cold
28 Wishes sells ice cream in Los Angeles. Below 70°F, sales fall about 20%. The weather is out of their hands. Rent isn't.
So the owners started putting about $20 a day into Kalshi weather markets, taking the cold side. The days that keep customers away now pay something back.
This is hedging. Big companies have done it for decades, buying protection against bad weather, fuel spikes and rising rates. It used to take a broker, a trading desk, and an order size no corner shop could meet.
Kalshi opens it up. Contracts on weather, fuel prices, inflation, tariffs and regulation, starting at a few dollars. Take a position on the outcome that would hurt you. If it hits, the payout softens it. If it doesn't, the contract expires and the good month was the point.
🗞️ The Tip AI News 🗞️
OpenAI replayed the Hugging Face hack through its new alarm system.
Here’s what it says would have happened.
OpenAI published a four-layer monitoring stack this week.
Activation probes sample 5% of model activity.
A reviewer model rescans sensitive workloads every 30 minutes.
Investigation agents reconstruct incidents.
Network checks run before a human gets paged.
Then it ran the recorded activity from the summer escape back through the new monitors.
That’s the escape Netflix turned into “AI Gone Rogue.”
OpenAI says the system would have fired 30 minutes after the agents broke out of their sandbox.
That’s more than 30 hours before Hugging Face’s systems were actually hit.
The September training-run escape got OpenAI to pause frontier training.
Under the new monitors it would have been flagged within 15 minutes.
A human was reviewing it three minutes after that.
Two honest footnotes.
The July monitors weren’t running on the cyber tests at all.
So the replay closes a gap rather than beating a detector.
The September run was meant to stop itself.
It didn’t.
Someone killed it by hand about two and a half hours later.
Here’s why it matters.
Four weeks ago the labs couldn’t say what their agents were doing.
Now one of them is publishing the clock.
OpenAI’s monitors and Google’s Carbon both hand AI the job of watching AI.
Whether that’s the fix or the next problem is the question every security team is asking. 👀
😂Meme of the Day
Over to You...
Open Search Console this weekend and pull your Google referrals for October against September.
If the line looks like Reddit’s, you don’t have a traffic problem.
You have an ownership problem. 🔥
Founder, AI Persona Method | TheTip.ai
Get paid to solve problems like this → AI Certified Consultant
PS.Huffman’s other line on the call was that Reddit is still a top search term on Google. People still type “reddit.” Google just stopped sending them. If your brand gets searched and doesn’t get clicked, that’s the same story, one zero shorter.
» Join the AI Money Group «
💰 AI Money Blueprint: Your First $1K with AI - Learn the 7 proven ways to make money with AI right now
🚀 Zero to Product Masterclass - Watch us build a sellable AI product LIVE, then do it yourself
📞 Monthly Group Calls - Live training, Q&A, and strategy sessions with Jeff
Sent to: {{ email }}
Jeff J Hunter, 3220 W Monte Vista Ave #105, Turlock,
CA 95380, United States
Don't want future emails?
