Riding the AI Wave: the Visionaries, the Hype Machines, and the AI Bros

As I sipped my coffee and scrolled through my Twitter feed this morning, a kaleidoscope of tools, opinions, hype, and experiments with AI vied for my attention—particularly around the breakneck pace of innovation happening with Large Language Models (LLMs).

The Rise of the AI Bro

Amidst the cacophony, I noticed a common thread: a significant portion of these Twitter accounts focused less on developing their own AI projects or experiments and more on stirring up enthusiasm for other’s work, casting predictions of the coming job-apocalypse, and generally aggregating news about AI rather than making new things.

These accounts could be classified as “AI Influencers” or, as I will refer to them in this article, the “AI Bros” of Twitter.

To be fair, there are some genuine experts and personalities on Twitter that I wouldn’t put in the same category as the AI Bros—folks like Lex Fridman, a research scientist at MIT and the host of the Lex Fridman Podcast, or Andrew Ng, the former head of Baidu AI Group and founder of Google Brain.

No, the prototypical “AI Bro” isn’t necessarily an AI expert, nor are they typically building anything with AI.

They are, however, quite adept at online marketing.

How To Know That You Are The Product

The phrase “if you’re not paying for the product, you are the product” is often attributed to Marc Andreessen, but the concept has been around since the seventies or eighties.

In commercial broadcasting the viewer pays for the privilege of having himself sold. You are the product of TV. You are delivered to the advertiser, who is the customer. He consumes you.

Annette Michelson—Richard Serra, Interviews, Etc. 1970-1980

Every YouTuber ever will ask you to “Smash that like button and ring the bell” because the more of their content you consume, the more advertisements can be served.

For a masterclass in marketing one’s self in the AI space, look no further than Rowan Cheung, founder of The Rundown.

Twitter Thread Marketing: Four Easy Steps

I like to call this technique “the threadicle” after the more common blogging phenomenon called “the listicle.” This isn’t unique to AI but is highly prevalent in the space and thus, likely very effective.

  • Step one: Hunt down any intriguing content others create within your chosen niche, sparing no effort.
  • Step two: Compose a detailed Twitter thread highlighting these exceptional projects and products.
  • Step three: Humbly request a follow and a retweet, emphasizing your desire to help others while expanding your audience.
  • Step four: Convert those precious eyeballs into a steady revenue stream.

There are many ways to monetize your audience: display ads, sponsored content, affiliate marketing, subscriptions—and there’s nothing wrong with making a buck! You’ll find monetization techniques right here on Bullshit Network (though we strive to leave out the bullshit).

AI Is Coming For Your Job

It’s already happening! Take a look at this story of Greg Isenberg’s totally real copywriter friend:

Be sure to like/reply/retweet.

Nothing builds an audience better than fear, uncertainty, and doubt, but at least Greg put a positive spin on this genuine, nonfictional anecdote. Turn lemons into lemonade!

I agree there’s a storm brewing, and AI will catch many people by surprise.

However, while GPT-4 and similar LLMs have the potential to automate various tasks and impact jobs, it’s essential to consider that they can also complement human expertise, leading to more efficient outcomes and new job opportunities.

Prompt Bros Are the Real AI Bros

Writing decent LLM prompts is a relatively easy skill to learn. These things churn out exciting content with the most basic-ass prompts you could imagine. Here is what GPT-4 gave me when I used the prompt “Cantaloupe”:

Cantaloupe, also known as muskmelon or Cucumis melo, is a type of fruit that belongs to the Cucurbitaceae family. It is closely related to other melons like honeydew, watermelon, and cucumbers. Cantaloupe is characterized by its round or oval shape, rough netted skin, and orange, sweet, juicy flesh. It is rich in vitamins A and C, potassium, and other essential nutrients.

GPT-4

Then I prompted: “Make it fantasy”:

In the magical realm of Fruitalia lies the mystical Cantaloupe of the Enchanted Grove, a fruit created by an ancient order of sorcerers called the Melonmancers. This extraordinary cantaloupe is believed to have powerful healing properties and grant temporary invulnerability to those who consume it. The enchanted fruit is protected by Melonbeasts, sentient cantaloupe-like creatures with fearsome, vine-like tendrils.

GPT-4

Then I added, “Make that into a tweet.”

🍈✨ In the magical realm of Fruitalia lies the mystical Cantaloupe of the Enchanted Grove! Created by Melonmancers, this extraordinary fruit is said to heal wounds & grant invulnerability. But beware, fearsome Melonbeasts protect it! 🐲 #FantasyFruit #FruitaliaAdventure

GPT-4

This isn’t rocket science. You don’t need to subscribe to a free newsletter to unlock the secrets of good prompting.

You just need to start prompting.

The Call Is Coming From Inside The House

Wait… is this a case of the pot calling the kettle black? Am I just another AI Bro? Well, maybe a little.

Screenshot of Jacquard AI interface.

But I don’t just sit on the sidelines aggregating content! I started building Jacquard AI, a friendly framework for building task-driven autonomous agents that certainly won’t escape and kill us all one day. It’s similar to BabyAGI, and AutoGPT, but it’s written in Go as a responsive CLI tool that aims to be friendly and easy to use.

It still needs to be finished because I’ve been too busy scrolling Twitter and writing this blog to write all the code. I’ll tweet more about it on my personal Twitter when it’s closer to done.

Are AI Bros Even a Problem?

Screenshot of Jacquard AI interface.

AI Bros aren’t necessarily problematic in the conventional sense. Their presence in the AI landscape is a natural consequence of the growing interest in the field.

Nevertheless, I would contend that, as public discourse zeroes in on concerns about AI potentially stealing the work of traditional artists, content aggregators are getting a free pass. They ride the wave of AI excitement, reaping the benefits of the hype without necessarily contributing to the field.

Anyone can be a hype man, and there is no shortage of people jumping on the AI bandwagon to capitalize on its popularity. But if someone is profiting from the hype, they must do more than just amplify the noise. Ideally, they should add value to the AI community by creating original AI content, developing AI tools, fostering communities, or in other ways that have a net positive impact on the ecosystem.

By encouraging a higher standard of contribution from AI influencers, the community can ensure that the discourse surrounding this transformative technology remains grounded in reality and focused on its true potential.

Above all, we must recognize and celebrate the individuals contributing to AI’s growth and understanding while fostering a culture of critical thinking and constructive engagement.