Matthias

Ex-Tesla | €100M+ in Paid Media Managed | Your ROAS is lying to you. Let’s fix that.

[Matthias @msteptb]

9 metrics are all your CMO dashboard needs.

https://t.co/Zp3d5qISLG

#marketing

[Melissa Chen @MsMelChen]

Singapore’s Foreign Minister, Dr Balakrishnan casually explaining how he built his own AI agent (a 2nd brain for diplomacy) using Claude & WhatsApp integration etc. on a Raspberry Pi

“You cannot govern a technology you have only been briefed on.” 🇸🇬

[BlueAlpha @BlueAlphaAI]

Install the BlueAlpha Marketing Plugin for Google Ads in Cowork or Claude Code, connect your accounts via MCP, and run your first live audit — all in under two minutes of setup.
The plugin packages ten agent skills that connect directly to the Google Ads API: account audit, budget reallocation, creative fatigue detection, competitive conquest, geo expansion, incrementality test design, and more. Each skill follows the same loop: Analyze the account, produce a recommendation with the data behind it, then execute approved changes on-platform.

This is what BlueAlpha calls the Decision Engine — the system that turns every signal in your ad account into a real decision. Not a dashboard. Not a reporting layer. A team of agents that does the analytical work so your growth lead can focus on the call, not the dig.

📌 What's covered:
0:00 – Intro
0:13 – Install in Cowork (drag-and-drop from GitHub)
1:00 – Install in Claude Code (two slash commands)
2:27 – Connecting the BlueAlpha MCP to your Google Ads accounts
3:30 – Running auto-optimize on a live account
5:04 – MCC support: running across multiple accounts
6:08 – Manager vs child account routing
6:40 – Audit results + markdown report output
7:06 – Other skills you can run

🔗 Links:
Full skill reference + copy-paste prompts → https://t.co/M9FIARE1HA
Setup documentation → https://t.co/NHSq7FYe0n
Plugin download (GitHub) → https://t.co/pzc4OVsSwY
Book a demo → https://t.co/ev759lnGrW

The BlueAlpha Marketing Plugin works inside Cowork and Claude Code. One MCP connector, one sign-in, no developer tokens, no OAuth configuration.

#GoogleAds #Claude #MCP #MarketingAutomation #BlueAlpha #GoogleAdsMCP #AIMarketing

[unusual_whales @unusual_whales]

Revenue generated per dollar of fixed assets, per FT:

[Hedgie @HedgieMarkets]

🦔A researcher invented a fake eye condition called bixonimania, uploaded two obviously fraudulent papers about it to an academic server, and watched major AI systems present it as real medicine within weeks.
The fake papers thanked Starfleet Academy, cited funding from the Professor Sideshow Bob Foundation and the University of Fellowship of the Ring, and stated mid-paper that the entire thing was made up. Google's Gemini told users it was caused by blue light. Perplexity cited its prevalence at one in 90,000 people.

ChatGPT advised users whether their symptoms matched. The fake research was then cited in a peer-reviewed journal that only retracted it after Nature contacted the publisher.

My Take
The researcher made the papers as obviously fake as possible on purpose. The AI systems didn't catch it. Neither did the human researchers who cited it in real journals, which means people are feeding AI-generated references into their work without reading what they're actually citing.

I've covered the FDA using AI for drug review, the NYC hospital CEO ready to replace radiologists, and ChatGPT Health launching this year. All of that is happening in the same environment where a condition funded by a Simpsons character and endorsed by the crew of the Enterprise was being presented as emerging medical consensus. The people making these deployment decisions seem to believe the pipeline from research to AI to patient is more supervised than it actually is. This experiment suggests it isn't supervised much at all.

Hedgie🤗
https://t.co/8Kg8FOrgHW

[Gaetano DiNardi @gaetano_nyc]

Here’s why chasing traffic ultimately fails. REAL-WORLD client example 👇

In late 2024, Crowdstrike had a massive outage that crashed millions of systems. It caused their stock price to crash by nearly 50%.

A client of mine saw this as an opportunity to “newsjack” by stealing their brand traffic.

So we built Crowdstrike alternative pages, pros and cons pages, etc.

We saw a huge surge of traffic almost immediately.

The big burst of traffic eventually normalized but we continued getting residual traffic over the years for Crowdstrike alternatives and other types of long-tail searches.

The problem though?

The client’s product DID NOT ALIGN with buyer intent for Crowdstrike alternatives.

They overlapped with one small feature of Crowdstrike, so ICP overlap was strong, but use case and intent was way off.

As a result, we saw extremely low ROI of “stealing” traffic from Crowdstrike.

And we never got recommended by AI answers.

This also created a bigger problem. OVERBLOATED organic traffic.

Thousands of clicks per month that were never going to drive demos.

Every month our “top pages” report was cluttered up with irrelevant Crowdstrike pages.

Two years later and we finally took those pages down.

So all that? For minimal ROI and a lot of noise.

“Stop chasing traffic” was good advice, even before AI.

[Hedgie @HedgieMarkets]

🦔Researchers at the University of Pennsylvania studied what they call cognitive surrender, the tendency to accept AI outputs without critical evaluation. Across 1,372 participants and over 9,500 trials, subjects accepted faulty AI reasoning 73.2% of the time and only overruled it 19.7% of the time. When the AI was wrong, users still accepted its answer 80% of the time. Subjects who used AI scored 11.7% higher on confidence in their answers despite the AI being wrong half the time. Adding time pressure made people 12 percentage points less likely to catch AI errors. Adding financial incentives and immediate feedback made them 19 points more likely to catch them.

My Take
The time pressure finding matters enormously for how AI is actually being deployed in workplaces. Companies are using AI to justify faster turnaround times, which means employees are using it under exactly the conditions that make them least likely to catch mistakes. When you're rushed, your internal monitor for detecting errors essentially stops firing, so you get AI output, no time to review it, high confidence it's correct, and a meaningful chance it's wrong.

People using a system that was wrong half the time still felt more confident in their answers than people who weren't using AI at all. That is a system actively making people worse at knowing what they don't know, which is one of the most dangerous things you can do to human judgment at scale. The companies pushing AI hardest into employee workflows should be reading this research carefully.

Hedgie🤗

Link to research for those interested: https://t.co/0iZz42132M

[JH Scherck @JHTScherck]

There's a belief among seed and series A (first time) founders that the right way to approach SEO is programmatic content scaling.

Not sure where it's coming from, but seeing it on calls more and more, like it's the easy, obvious path to accelerate growth... and buddy — it's not

[Simon Kuestenmacher @simongerman600]

200 years ago China was the world's biggest manufacturer. The Brits overtook them just to be overtaken by the US shortly after. 20 years ago China once again climbed on rank 1. Source: https://t.co/UZ15S3OGTR

[unusual_whales @unusual_whales]

I want you to understand how unusual this is.

6:49 AM on Monday: A sudden surge in oil futures trading, approximately $580 million in volume, no news, no announcement, nothing public.

7:05 AM: Trump announces a pause on Iran strikes. Markets move immediately.

Roughly $580 million in contracts were placed about 16 minutes ahead of the news.

Unusual.

[Bubblemaps @bubblemaps]

BREAKING: 🇺🇸 🇮🇷 🇮🇱 Someone made ~$1,000,000 betting on US & Israeli strikes for the last TWO YEARS with near-perfect accuracy using 7 connected Polymarket accounts 🧵

[Google Search Central @googlesearchc]

Today we released the March 2026 spam update. We'll update our ranking release history page when the rollout is complete: https://t.co/CwCjstgv0H

[Jay Van Bavel, PhD @jayvanbavel]

There are big political and intergeneratonal differences in what people think is morally wrong

For example, Gen Z are 3X more likely than Boomers to see being extremely rich and 2X more likely to see spanking children as morally wrong

https://t.co/mOpwWnmoyo

[Tom Goodwin @tomfgoodwin]

Gemini has been lying to me for the last 30 mins about the idea it can pull out live flight price information.

It refuses to accept it can't access flight info.
It continually makes up prices and empty seats.
Its wild.

[Matthias @msteptb]

@gaetano_nyc Happens with migration to Framer as well!

[Lily Ray 😏 @lilyraynyc]

One of the worst things you can do for AI search visibility?

Destroying your SEO performance with shiny new AI search tactics that are ultimately dangerous for SEO - and by extension - AI search.

In my latest Substack article, I wrote about the fundamental connection between SEO and GEO/AEO, along with some of the risky tactics I've seen sites employing, plus the early signs that these approaches are dangerous for SEO (and therefore AI search as well).

https://t.co/2x5YLkRRvQ

[Jay Van Bavel, PhD @jayvanbavel]

LLMs overemphasize moral concerns common in Western societies and underestimate values more prominent elsewhere. These distortions likely stem from cultural biases in training data and carry societal implications and risks

https://t.co/0nacfBcoA7

[Matthias @msteptb]

Starting July 1, advertisers buying ad space in France, Austria, Italy, Spain, Turkey, and the UK will see a new line item on their Meta invoices: a surcharge matching (percentage for percentage: 2% to 5% depending on the country) the local Digital Services Tax rate.

The platforms were never going to pay the tax. The businesses absorbing these costs will be European SMEs, retailers, and service companies.

Something most people don't know: Google already did this quietly in 2021. Meta is doing it 'loudly', with a line item that makes sure every advertiser knows exactly who to blame. The likely consequence is that other platforms will follow. Why shouldn't Amazon, Spotify, LinkedIn, TikTok do the same?

European governments wanting to tax dominant digital companies is a legitimate objective. Why? Because they earn returns that are structurally protected by network effects, data lock-in, and market concentration rather than by ongoing competitive merit. Those supernormal returns are, in standard public finance terms, rents. And taxing rents does not distort productive behavior the way taxing normal returns does - it can actually make markets more competitive by reducing the payoff for entrenching monopoly power.

But DSTs can't achieve anything. A gross revenue tax on advertising touches the advertisers, not the platforms' margins.

Some tools that could actually work: enforcing the global minimum tax deal that 140 countries already signed, updating century-old rules that define when a company is 'present' enough in a country to be taxed there, designing levies that target excess profits specifically (rather than revenues).

Yes, these are harder to legislate and require a lot of coordination, but isn't that why we pay politicians? Rather than to craft easy announcements whose bill European advertisers end up paying. 🙄

[Matthias @msteptb]

@simongerman600 This basically means most people never switch from passive to active listening (actively searching for more/new music), as actively researching and listening, for the ones who do that, 1) definitely doesn't peak at that age, 2) keeps the influence window open

[Aleyda Solis 🕊️ @aleyda]

🚨 Google is responsible for nearly 3/4ths of all desktop web searches; ChatGPT is smaller than you might think - A new study from @randfish showing how:

* Google’s 2025 US market share is closer to ~70% than 90% when platforms beyond traditional search engines are included.
* Most AI Search and AI Answers happen on Google: Even if you combine every prompt on ChatGPT, Claude, Deepseek, and the rest and assume every prompt is a search-equivalent, Google dwarfs them.
* Amazon, Bing, and YouTube still receive more desktop search activity than ChatGPT, despite the latter’s buzz. If you’re worried about AIO/SEO-for-AI/GIEIO, you should be worried about search visibility in those places, too (if their audiences are relevant to your business).

Much more! Check it out: https://t.co/XFvRInB4sA

[Aleyda Solis 🕊️ @aleyda]

🤖 The real risk of AI-generated content - A must read article by @TomekRudzki for those using AI content generation tools in autopilot and with little to no quality assurance and editorial validation!

There are certainly relevant scenarios to use AI to enrich your content at scale for Websites that due to their model actually need it (as they need programmatic too) and do it while keeping the quality high, for which it's critical to establish a content quality assurance workflow that along with a high quality input, and an editorial validation and insights enrichment to ensure real, unique value and insights will actually make your content better... however, if you use AI content to replace what it should be expert or experienced content: It's a recipe for failure.

Take a look at the article: https://t.co/ouNQfcskFS