Reddit tells Big Tech to PAY UP! Musk hopes to "seek the truth". AI Drake = CANCELLED.
GM, and happy Thursday! Welcome back to another AIcyclopedia newsletter.
Here’s what we’ve got for ya today:
You want our data? Reddit says PAY UP!
Elon: May the TRUTH be with you
AI image wins award, but recipient refuses it to prove a point
AI Drake = CANCELLED
AI News: The Headliners 🏁
1/ You want our data? Reddit says PAY UP!
Reddit is adding another stream of revenue, and wouldn’t you know it… it’s in the AI space.
[The News] We all know Reddit. The place where people go to share, converse, and debate on anything and everything imaginable (and more often than not, come across a crazy video or two).
Let’s be real: It’s a treasure trove of data. And Big Tech AI companies want it.
But Reddit isn’t giving in so easily and is now taking advantage of the opportunity during the AI craze to monetize its gold mine of human interactions.
How it’s going to work is still pending.
But from what we know, Reddit is creating a “new premium access point for third parties,” in which companies can use the data from Reddit to create and train chatbots.
[Why this matters] A few months ago, Microsoft penalized and called out 3rd party search engines that were breaching contract by utilizing its proprietary search engine data to create chatbots of their own.
Similarly, Reddit is realizing that people will take advantage of its data and train AI models if they do not take control over who is able to access it.
Steve Huffman, CEO of Reddit, told The New York Times, “The Reddit corpus of data is really valuable. But we don’t need to give all of that value to some of the largest companies in the world for free.”
Looking forward… The power of Reddit’s information is its ability to incorporate person-to-person-like interactions into AI models.
If AI models are to continue to get better, then this type of personified data is important alongside factual data.
****
2/ Elon: May the TRUTH be with you
Once an investor and co-founder, Musk criticized OpenAI for “training AI to lie.”
Now, with his billions, he is creating his own chatbot to rival ChatGPT and Google.
But there’s a twist.
[The News] In an interview with Tucker Carlson on Fox News, Musk stated that “I'm going to start something which I call 'TruthGPT', or a maximum truth-seeking AI that tries to understand the nature of the universe.”
He also admits that this might be a little late as far as AI chatbots go, but he says that “TruthGPT might be the best path to safety that would be unlikely to annihilate humans.”
Essentially…he’s talking about a chatbot with self-awareness.
[Why this matters] Since OpenAI has become a for-profit entity, Musk is concerned that ChatGPT may interfere with the ethics of the AI models it develops.
$$$ + politics > Truth
TruthGPT, on the other hand, would do the opposite and become the truthful, unbiased, and transparent alternative to ChatGPT.
Looking forward… Musk has already established a new AI company called X.AI.
Does that mean TruthGPT is coming soon?
Maybe, maybe not. Who knows how it’s going to work out in the future?
But given Musk’s ability to realize the unimaginable despite the criticism he is already facing…
It doesn’t seem like he is fazed in the slightest.
Microsoft. Google. You better watch out.
Want to sponsor a future newsletter and share your business with our growing list of followers?
Go ahead and send an email to [daniel@aicyclopedia.com].
This space could be yours…🥳
AI News: The Shorts 🩳
3/ AI image wins award, but recipient refuses it to prove a point
[The News] CNN reported that Boris Eldagsen, based in Berlin, won an award for the 2023 Sony World Photography Award.
But here’s the catch: The image was generated using AI. (notice the hands??)
According to Eldagsen, he did initially notify the organizers that he’d incorporate AI into parts of his process while developing the image…
But did not state the final image was fully generated using AI.
The ultimate goal was to mislead the organization to show that AI-generated images should be separate from human-captured photography, as the organizers and judges were not able to tell the difference.
[Why this matters] Eldagsen told CNN: "It shows that at the moment the photographic world has been taken by surprise after this development that subtly you can create images that look like photography, but you don't need to have the skills and expertise of photographers."
So in a future global competition like this, anyone could fake it and earn notoriety for work that required little to no skill.
The end goal here was to create a conversation around AI and photography, as many photographers were left feeling “threatened and afraid that they are going to lose their jobs, which will happen."
****
4/ AI Drake = CANCELLED
If you’ve been on TikTok lately, then you've probably heard about the viral song “Heart on My Sleeve,” featuring Drake and The Weeknd.
The problem? It was AI-generated. (the AI used is unknown)
And the Universal Music Group is not having it.
[The News] On Tuesday morning, the viral AI-generated hit that hit 6.9 million views in a matter of days was removed from all music streaming services.
According to the anonymous musician who goes by "ghostwriter," he stated on TikTok that the musician wrote the song and then replaced their voice with Drake’s.
[Why this matters] UMG announced last week that it would begin cracking down on AI-generated music.
Not only is this to stop AI developers from training AI on copyrighted music, but it is also to protect the intellectual property of its artists.
Looking forward… Right now, the verdict is that the song infringes on Drake and The Weeknd’s right of publicity, as stated by Edward Klaris, a media lawyer.
But as for using copyrighted songs to train AI models, that is still up in the air and will need to be decided in court.
Want to hear the song? Just provide your phone number.
Your AI news break 🥳
AI News Tid Bits…
👉🏻 Microsoft announces development of its own AI chip, Athena. The hope is that it will be more powerful than the options currently available and will be used to power chatbots like ChatGPT. Nvidia? Bye, bye.
👉🏻 EU lawmakers are worried about the exponential development of AI and are calling a summit to discuss how it should be controlled. And yes… Biden is invited.
👉🏻 Before, it was Google vs ChatGPT. Now, Google is leveling up and developing a new AI-powered search engine. Competition? Google says “nahhh”.
👉🏻 The US FTC announced that the agency would “pursue companies who misuse artificial intelligence to violate laws against discrimination or be deceptive.”
Try this AI tool 🪄 : Hypertype AI
To anyone who works with customers or is in sales, this one’s for you.
You know it: The pain of wasting precious time answering the same ole’ email questions over and over again…
Or scrolling through 100s of past emails just to find that ONE piece of missing information you are looking for… (lunch break? what lunch break?)
And then manually answering every email one-by-one in the hopes that all the information is accurate and helpful.
#stressful #headache #afternoonmargs
But with Hypertype AI, you can scratch all of that ‘cause now, you’ll be #winning and #onarollbaby.
With your personal AI email writer, Hypertype AI chrome extension will:
Scan, compile, and keep track of your email data and history…
Fetch the most relevant information for your working emails…
Fill it in for you…
…all with the click of a button…
So you spend just a little bit of time doing manual email work and more time doing work that you care about.
(eh hem…like making that $$$)
[Here’s what people are saying]:
I've been working in sales for more than 10 years now, and I have been constantly in search of a magic instrument that could help me to change the routine of answering emails. Hypertype is awesome! It saves ton of time and energy, and does it in a very simple and elegant manner!
A very powerful extension! It's like having GPT, Grammarly, and search feature on steroids in one.
I've been following Hypertype for a while, and I really look forward to them helping me cut my email-writing time in half without losing the contextual relevance.
[See it in action]:
That’s it for today!
If you liked the newsletter, give it a ❤️ and, as always, feel free to sound off in the comment section below 👇🏾
Thank you for choosing us as your AI news source, and don’t forget to follow us on Twitter!
See you Sunday 👋🏻
