😺 AI detectors EPIC fail

Welcome, humans.

Reddit has declared the dawn of "The Facebook AI video slop era." Case in point? A clearly fake "giraffe rescue" video shared and liked by 23K people:

Or the viral image of a little girl clutching a puppy after Hurricane Helene—real or not, it's viral.

Some say it's bots all the way down now—bots creating content for bots, liked by bots, with advertisers footing the bill.

We know Zuck's all-in on millions of AI assistants, but if this keeps up, won't Facebook basically become the embodiment of the dead internet theory?

Here’s what you need to know about AI today:

• We break down the flaw in AI detectors, and how to write more human.
• DeepMind team won the Nobel Prize for AlphaFold.
• Amazon added new AI features to its delivery and shopping systems.
• OpenAI projected $44B in losses until 2029.

Advertise in The Neuron here.

AI detectors = not so smart after all?

Christopher Penn, co-founder and Chief Data Scientist at TrustInsights.ai, just dropped a truth bomb about AI detectors on LinkedIn. Spoiler alert: they're not as clever as you might’ve thought.

Here's the scoop: Penn ran the U.S. Declaration of Independence through an AI detector (you know, tools like ZeroGPT that claim to sniff out AI-generated content).

The result? ZeroGPT said the Declaration of Independence had a 97% chance of being AI-generated. Wait, what? Did Thomas Jefferson have a time machine and a ChatGPT subscription?

Nope. Turns out, AI detectors aren't exactly rocket science. Here's why they goofed:

1. They use metrics like “perplexity” (how predictable the text is) and “burstiness” (how much the writing style varies). The Declaration's precise language and similar sentence lengths made it look suspiciously AI-like.
2. These detectors are trained on vast amounts of text... including historical documents. So when they see the Declaration, they're like, “Hey, I've seen this before! Must be AI!”

The detector bots were like, “You might not be Nicolas Cage in National Treasure, but you DEF stole that my guy!”

Penn's verdict? “AI detectors are worthless.” Ouch.

So what's the solution? Penn suggests we humans need to prove our work's authenticity, kind of like how fancy cheeses come with certificates of origin.

Funny enough, on Tuesday, Adobe announced an app coming out next year that lets creators add that kind of digital signature to their work to prove it's human-made.

Our take: If the Founding Fathers can be mistaken for AI, we're all in trouble. Want to sound more human in your writing (whether you're using AI or not)? Try these tips:

1. Mix up your sentence lengths. Short ones. Longer, more complex ones. You get the idea.
2. Throw in some unexpected phrases… like comparing AI detectors to cheese certificates!
3. Add a personal touch. AI probably won't mention that time you spilled oatmeal on your J Crew tee while writing about AI…

Remember, the goal isn't to trick AI detectors—it's to communicate effectively and authentically.

FROM OUR PARTNERS

Want to dominate sales in 2024? Attention's AI assistants are your ticket to the top.

While other sales teams are drowning in follow-ups and data entry, smart companies are leveraging Attention's AI to stay ahead of the game:

• Send personalized follow-up emails within minutes of ending a call.
• Get cross-call insights to understand why you're winning (or losing) deals.
• Receive live coaching during calls to handle objections like a pro.

Attention’s users are seeing incredible results:

• “Game-changer for our productivity. Thanks to Attention, our reps save 20-30 minutes per call with automated CRM data entry” - Yan Kessler, VP Sales at Aspire Technologies.

Never lose a deal to slow follow-up again. Automate your follow-up game and close more deals, faster than ever before with Attention.

Around the Horn.

More crazy NVIDIA performance details here.

• Another AI win—Google DeepMind's Demis Hassabis and John Jumper won the Nobel Prize in Chemistry for AlphaFold, their AI that predicts protein shapes.
• Amazon’s new warehouse will employ 10x more robots, its EV fleet will get AI-powered cameras to show drivers which boxes to grab at each stop, and its website got AI shopping guides for hundreds of product categories—oh yeah, and it’ll eventually get AI agents that add items to your cart for you.
• Writer, an AI text generator for corporate clients, trained a new model for only $700K using synthetic data—and now its fundraising at a $1.9B valuation to compete with OpenAI, Anthropic, and Cohere in the enterprise market.
• Internal docs revealed OpenAI expects to lose $14B in 2026, projects total losses of about $44B total by 2028, and says it won’t turn a profit until 2029—”when” its revenue hits $100B. Oh, and it’s becoming a public benefit corporation specifically to prevent a hostile takeover from Microsoft.

Treats To Try.

1. *Want to close your B2B deals faster without compromising on price? DIVACS' platform quantifies your product's real impact with your prospect or customer, making your value crystal clear for buyers. Request a Demo today.
2. Marqo (from former Amazon engineers) helps you create and use smarter search systems that understand the meaning behind words and images.
3. FinetuneDB allows you to build custom AI models with your data in minutes (newly launched, free to try, read more about it here).
4. OpenBB helps you research investments faster by combining data and AI analysis (we messed around w/ it, watch this for more info).
5. Arcads creats video ads with pretty convincing AI actors (demo vid here—we’ve seen some crazy ads made with this, but it costs ~$100 FYI).
6. Miro 2.0 now transforms your team's sticky note ideas into structured documents, turns canvas sketches into clickable prototypes, and suggests improvements throughout your projects (demo video here).
7. CortexSheet integrates ChatGPT into Google Sheets, allowing you to automate tasks like data cleaning, content generation, and formula creation directly in your spreadsheets (demo video here—free trial avail, then $8 a month FYI).

See our top 51 AI Tools for Business here!

*This is sponsored content. Advertise in The Neuron here.

Thursday Ticker.

Since most ppl still haven’t tried it, here’s another demo of Canvas for coding.

• W.S. chose Prefer it over regular GPT: “I loved being able to edit in the Canvas and ask GPT to make changes too. It felt like we were working on my copy together. Plus, it reduces the length of the thread – where it used to spit out the whole thing again every time I asked for changes.”
• B.J. chose Haven’t used it yet: “For some artifacts Claude already has this feature, not as advanced as the new Canvas feature. But it was there before I heard about Canvas ;-)”
• L.K. chose Prefer it over regular GPT (with a twist): “…there are different use cases for each. In fact, I'm finding that I use ChatGPT voice on mobile to help get an idea formulated, then I'll move to Canvas to edit. That's a real time saver for me! I also use it for longer, in depth copy so that I can move around and edit as it develops.”

A Cat's Commentary.