AI: Video hits Home

...as Meta shows, with Advertisers & Users

The expression ‘To a hammer everything is a nail’ comes to mind only nine months after the world was electrified by OpenAI’s ChatGPT. Since then, we think of LLM AI software as a source for magical ‘Sparks’, answers and perspectives, driven by smartly engineered text “Prompts”.

We’re going to access AI as much via Voice soon as with text prompts. Google, Apple, Amazon and others are hard at work adding LLM AI capabilities to Google Assistant, Siri, Alexa and more.

So not everything AI has to look like OpenAI’s ChatGPT as it’s used today. We’re going to see LLM AI products, services and applications in a dizzying array of forms. Many different nails for the AI Hammer. 

And some of the early, big AI driven revenue dollars are showing up in Video.

A vivid example of that was Meta’s quarter just reported, where Founder/CEO Mark Zuckerberg walked through the range of AI driven product opportunities for Meta (links mine):

“We continue to see strong engagement across our apps and we have the most exciting roadmap I've seen in a while with Llama 2, Threads, Reels, new AI products in the pipeline, and the launch of [VR product] Quest 3 this fall,"

“Meta's second-quarter revenue grew 11% to $32 billion in the quarter ended June 30”, ahead of analyst expectations.”

As I’ve highlighted before, for companies like Meta and Google, who are the market leaders in online Ad Sales, the biggest near-term opportunity is generating large, accelerating revenue streams with AI, on the Advertising side first. We saw that in spades this quarter:

“Reels, Meta Platforms’  answer to viral short-form video app TikTok, elicited eye rolls when it launched in 2020 and was regarded as yet another example of Meta copying a popular rival.”

“But on Wednesday, Meta revealed numbers that show Reels videos are growing rapidly among both users and advertisers and are quickly catching up to the ByteDance-owned TikTok app that is beloved by young users and has reshaped the social media landscape.”

“Meta says the rags-to-riches growth of Reels, which are viewable on both Facebook and Instagram, speaks in part to the company's improving recommendation software, long the Chinese-owned rival's strong point.”

"We can show Reels that we think you're interested in based on our discovery engines," Justin Osofsky, Meta's head of online sales, operations and partnerships, said in an interview.”

“The number of Reels video plays on Facebook and Instagram now top 200 billion per day, up from 140 billion last fall.”

“Though Meta has invested in artificial intelligence for years, it is now moving to the forefront of its business to improve content recommendations and ads across the company's services, he added.”

Here’s the punchline on the breathtaking momentum of AI driven ad sales vs TikTok (Bold text mine):

“Reels' annual revenue run rate has jumped to $10 billion, up from around $3 billion as of last fall and $1 billion last summer, Meta CEO Mark Zuckerberg said during a conference call with analysts following its second quarter results.”

“That means Reels is about the size of TikTok's business as of last year in 2022, when it pulled in $9.9 billion in worldwide ad revenue, according to estimates from research firm Insider Intelligence, which forecast $13.2 billion in global ad revenue this year.”

For advertisers, it’s one stop convenience, at scale:

“One reason for Reels' growth is that Meta's ad platform makes it seamless for advertisers to place their promotions on the feature, said Debra Aho Williamson, a principal analyst at Insider Intelligence.”

"It's as easy as checking a box," she said.”

More than three-quarters of Meta's advertisers are placing ads on Reels, said Susan Li, Meta's chief financial officer.”

This despite the reality that Meta/Instagram Reels is not seen by many to be as good as TikTok either in the amount of video content available, or in the quality of the AI “For You” algorithms that match that video to user tastes. 

Note this video discussion by Nilay Patel, the editor at The Verge, and his colleagues comparing Reels and TikTok, with Google YouTube Shorts in the mix. Click to the 13 to 16 minute mark for a really telling candid discussion on this subject. 

The reason I am emphasizing this topic and the numbers, is that video is one of the primary near-term daily AI drivers for billions of users daily, whether it’s on TikTok, Meta, Google, Netflix, Elon’s ‘X-Twitter’,  or any number of online services. TikTok’s “For You” AI driven video feeds are the model for the industry, and everyone is racing down that path at scale. 

As I’ve also outlined separately, video is likely to be the next huge amount of ‘Extractive Data’ for Foundation LLM AI models in the next few years, fundamentally improving what LLM AI can do for mainstream audiences worldwide. As the Information explains:

“Google is upgrading its Bard chatbot with a new machine-learning model that can better understand conversational language and compete with OpenAI’s ChatGPT. As Google develops a sequel to that model, it may hold a trump card: YouTube.”

“The video site, which Google owns, is the single biggest and richest source of imagery, audio and text transcripts on the internet. And Google’s researchers have been using YouTube to develop its next large-language model, Gemini, according to a person with knowledge of the situation. The value of YouTube hasn’t been lost on OpenAI, either: The startup has secretly used data from the site to train some of its artificial intelligence models, said one person with direct knowledge of the effort.”

“AI practitioners who compete with Google say the company may gain an edge from owning YouTube, which gives it more complete access to the video data than rivals that scrape the videos. That’s especially important as AI developers face new obstacles to finding high-quality data on which to train and improve their models.”

Google, Meta and TikTok are obviously at the head of this list to make this happen, both in terms of new AI video driven data and content, and Advertisers signed on at scale. And they are spending billions on Nvidia GPU related Cloud Capex to get there. 

As we saw from this quarter, Meta already, and likely Google soon, are already at the head of the line monetizing that AI around Videos with advertisers for now. And the numbers are getting real. Even before AI really starts to show up in so many other forms beyond ChatGPT. Voice and Video will lead the way. Stay tuned.

(NOTE: The discussions here are for information purposes only, and not meant as investment advice at any time. Thanks for joining us here).