Journalists Are Wetting The Bed

Bye Bye Journalism ✍️

With every passing day, we are watching old-school press fade away.

The rise of citizen journalism and AI’s ability work up stories is quickly leaving mainstream media in the dust.

Will we still be following people for the news or will a Google-trained bot report on today’s trending topics?

Today’s Menu

• Google Genesis: AI Owns The News

• AI Can’t Tell What’s AI: Silly Robot

• Meme: Meme

Read Time: 2 minutes

Code Name: Genesis 🤫

ChatGPT was never really an issue for journalists because it simply can’t do what a journalist does.

It’s a glorified word organizer. Not a truth collector or an imaginative storyteller.

But Google Genesis is a different story.

Genesis, as it’s currently understood, can quickly piece together something resembling a news article based off real-time data.

If Ryan Gosling said something on the Barbie press tour, and it was uploaded it to the Internet, Google Genesis would be able to whip up a trending article in seconds.

And who do you think would win in the search results?

It won’t be long until The Google Times, The Microsoft Post, and OpenAI News controls the narrative.

From our sponsor:

Find The Right Freelancer In Minutes 🤖

Fiverr connects businesses with talented freelancers from around the world, and offers digital services in over 500 categories.

• Get Quality Work Done Fast

• Stick To Your Budget

• Pay When You’re Satisfied

The Big Bite
AI Can’t Tell What’s AI 🔍

Large language models and AI chatbots are flooding the internet with auto-generated text, and there’s nothing we can do to stop it.

“Dude, no problemo. Just create a detection system.”

Yeah… you see, that IS the problem.

We already tried that.

It’s becoming so hard to distinguish AI-generated text from human writing that even OpenAI shut down their detection system.

It. Didn’t. Work.

Every day a new AI service appears that can create convincing and useful written content, and like always, this is being used for good and bad things.

Writing code? Yeah, a lot easier.

Writing factual errors and lies? Yup, that happens too.

If OpenAI can’t spot AI-generated content, how can anyone else?

Being unable to tell the difference between AI and human text leads to a problematic online world.

AI needs to be trained on good, human data, and if we can’t tell what “human data” is, LLMs may be trained on other LLM data.

When this happens, the whole system folds in on itself like a blackhole, called a “Model Collapse.”

If you want to save the world, I guess you’re going to have to start writing your own content again. 🤷‍♂️

Today’s Prompt Spotlight 🏆

Apply for tomorrow’s spotlight.

Fill out the application & submit your best prompt.

One Day In The Future… 🔮

This will be how AI delivers your news:

@localrandomtiktoker

Interesting way to make popcorn tho #live #tiktok #fyp #xyzbca

Today’s Fast Snacks 🥡

How Is AI Using Your Data?

They could be selling your information on the black market… or maybe they’re using it to improve their systems.

Politicians vs. Artificial Intelligence

AI could reboot democracy and improve our information ecosystems. Will you vote for an AI-powered politician?

Advertise With Snack Prompt

Our newsletter is read by thousands of tech professionals, investors, engineers, and business owners around the world. Learn more or book your spot today...

Reader Review Of The Day