When data becomes dangerous

I wish I didn’t have to address this. 

But in the aftermath of ICE protestor Alex Pretti’s death, I can’t help but think about the role data will play as constitutional rights are at risk.

I was on a hike this weekend through the jungle of Belize with a mentor of mine who had lived through the civil rights movement in Mississippi. 

She reminded me how much people who spoke out and refused to enforce segregation laws were equally punished. 

For example, when the Jewish community spoke up and defended blacks, the Ku Klux Klan bombed their synagogues.

Like most millennials, I’ve freely shared and expressed myself online since those early days of AOL instant messaging.

That love of expression continued through my journalism career.

I’ve given access to my personal data without much of a second thought, trusting in a free society.

But now as first amendment rights also become threatened, I’m thinking more about how data could be used in the future. 

Whether it’s our user data on apps or what we share online.

As an American citizen, I’ve often taken my expression for granted.

Now is not the time to shrink. I admire so many of the citizen journalists showing us what’s really going on in Minnesota and the firsthand videos and testimonies that challenge official accounts.

 If you are asking yourself what to do right now, I would suggest starting a newsletter or a platform to share what’s important to you or amplifying the stories of those who need the world to listen and hear the truth. 

Today’s issue of HyperFix dives into how you can do just that.

Gone are the days you could have AI write your emails, social media posts and reports without being detected.

This Reddit post went viral recently, highlighting the dead giveaways that you’re using AI to write content for you. It catalogs everything from em dashes to repetitive phrases like “let me break this down for you,” and “here’s the truth.”

This thread is especially helpful for training LLMs as well as your own eye. It also appears that ChatGPT is the culprit of much of this style, which is why it’s important to experiment with other tools such as Claude and Gemini.

When the reader detects your post is written by AI you run the risk of losing trust and confidence. Increasingly more creators are crediting ChatGPT or using tags like “Written by me with help from ChatGPT.”

People are deleting their TikTok accounts. Why? The Chinese-owned social media company struck a deal with U.S. investors that gives Oracle control over a U.S. version of the app.

Many users have reported deleting their accounts due to concerns about censorship and data security. Many content creators who receive income from the platform reported issues with uploading videos over the weekend and engagement rates plummeting to zero views.

TikTok’s U.S. app also broadens data collection with sensitive categories like citizenship and expanding location tracking.

My take: This is another case for building a media asset you own such as email newsletters. While email newsletters rely on email service providers and deliverability can be impacted, you will always have full control over your list of subscribers. That means you can take them with you wherever you go. 

Japanese researchers have found a way to reduce AI’s energy consumption by using real world traffic as a computing system. It uses traffic flow on road networks as the computational reservoir. The system processes data through the constantly changing interactions between vehicles. What they found is that this method can integrate directly into physical computing systems and reduce the overall energy demand. 

🗞️NEWSLETTER BIZ LAB
The New Media Summit is next month in Austin, Texas, and I’d love to see you there. This live two-day event is hosted by GrowLetter founder Matt McGarry and features the top new media and content economy creators who are turning their own media platforms, newsletters, podcasts into profitable brands. I’m especially excited to hear from Sam Parr, Codie Sanchez and Ryan Deiss. If you’d like to join, snag your ticket here.

🔥 EXTRA HYPE

• Singapore & UAE lead AI adoption rates worldwide 

• Gemini teaches students how to write with AI tutor 

• Vibe-coding platform Loveable doubles revenue to $200 million in 4 months

• High-profile Social Media addiction case against Meta begins 

• How to get hundreds of email subscribers from one YouTube video

• Anthropic CEO Dario Amodei addresses Claude’s blackmail problem & other AI risks

With curiosity,
Lucy Elizabeth, cofounder ThinkSpiral

📨 P.S. If you are building a newsletter and want to eliminate the weekly scramble, that is exactly what we help our customers do.

 👉 Book a consult with our team

And if this sparked something, forward it to an AI-curious friend.