This Week in AI News: October 11, 2024

Human Driven AI is officially launching our weekly AI News Brief Video Series, featuring our AI avatar, Zara Quinn, whom we introduced earlier this week. This new video series will spotlight the AI news stories that catch our attention each week. Today, it’s Canva’s price hike u-turn, Meta’s Ray-Ban hack, OpenAI’s plans for SearchGPT, Google’s and Meta’s hope for AI-generated Holiday ads, Zoom’s plan to clone us all, Google’s new ads in AI Overviews, and how ChatGPT is being used to spread election disinformation.

You can watch the first video in this series here or subscribe to our YouTube channel for this and a lot of other content. A full transcript of the AI News Brief is below.

TRANSCRIPT:

Hey there, AI enthusiasts and marketers! It’s Zara Quinn, coming at you with this week’s AI Brief. Buckle up, because it’s been a week of digital drama that’ll make your Roomba look like a push broom with ambition.

Canva’s Price Hike U-Turn

First up, a few months ago, Canva decided to pull a classic “oops, my bad” after trying to sneak in a 300% price hike that went over about as well as a pop-up ad at a web designer convention.

They’ve reversed the planned price hike and are now promising to keep things “fair and transparent” with their new “Pricing Promise,” including a pledge to offer a range of affordable plans for everyone with transparent and timely communications and no hidden fees.

We don’t yet know if this will be enough to bring back the customers who left the platform when the price hike was announced. We will let you know how this shakes out in the coming months.

Meta’s Ray-Bans: Now With Extra Creepy Features!

This week, two Harvard students hacked Meta’s AI Ray-Bans adding facial recognition technology that instantly identifies strangers in public, finds their personal information, including name, address, phone number, relative’s names, recent activities, memberships and more, all of which and can be used to approach them and gain their trust.

Great! Now strangers can know everything about you before you even say “hello.” It’s like stranger danger, but the stranger already knows your life story and your dog’s middle name.

This further illustrates the need for significance AI regulations, governance and guardrails.

SearchGPT: Google’s Newest Frenemy

Meanwhile, OpenAI is gearing up to launch SearchGPT, threatening Google’s search monopoly like a toddler threatens a sumo wrestler. SearchGPT will go live by the end of the year.

According to, OpenAI’s head of media, the new functionality will be integrated into ChatGPT, allowing users to quickly get answers to queries without needing to click through to other websites.

SearchGPT is still in testing, and early reports suggest it may need improvements before competing with major search engines like Google. However, OpenAI’s push to challenge Google’s dominance could be a game-changer in the battle for search engine supremacy.

The company says SearchGPT will offer natural language responses and provide sources for users who prefer to read original content, striking a balance between quick answers and reliable information. Time will tell if OpenAI’s latest product can truly rival Google.

AI-Generated Holiday Ads: Where Algorithms and Eggnog Finally Meet

Meta and Google are pushing for AI-generated holiday ads. Because if there’s one thing that captures the warmth and joy of the season, it’s an algorithm’s idea of what humans find festive.

The two companies are encouraging marketers to use generative AI for holiday ads, but agencies and brands are still hesitant. Meta’s Advantage+ ad platform suggests using AI to change backgrounds, resize images, and generate different versions of ad text. Google also recommends using AI to create new holiday imagery and automate ad headlines.

Advertising professionals have been using Generative AI in some parts of their creative process, like brainstorming and copywriting. But, most aren’t ready to cede control of the creative process to AI just yet.

We anticipate brands with fewer resources may find AI tools helpful to produce the high volume of ads needed for the holiday season. It will be interesting to see how AI-generated advertising plays out during the holiday season.

Zoom’s New AI Avatars: Because Even Your Digital Clone is Better at Meetings Than You

Zoom just announced they’re cooking up AI avatars for 2025 that’ll look and sound just like you.

Yeah, you heard that right – your very own digital clone to send to all those meetings you’d rather skip.

Just record yourself talking, let the AI work its magic, and you’ve got a mini-me ready to nod along to quarterly reports while you’re busy “focusing on actual work” (aka binge-watching that new series).

Google’s AI Overviews Now Include Ads Because Even AI Has to Pay the Bills

So, I’m sure you’ve seen the AI Overviews in your Search results lately. Google calls it Search Generative Experience and they’ve just allowed paid content to be included in these AI Overviews.

I’m telling you, Google is rolling out this update faster than a guy ghosting his Tinder date. Why? Well, it seems Perplexity is stealing their lunch money by providing people with answers to many of the questions traditionally thrown at Google. The search giant has already lost around 12% of search activity to AI. So, they are fighting back.

But don’t worry, Google’s AI is totally reliable. It only occasionally tells you to put glue on your pizza or cites The Onion as a credible source. It’s like that one friend who’s always ‘heard something’ but can never remember where.The ads, which have a “Sponsored” label, will appear alongside other, non-sponsored content in the AI summaries and will be drawn from advertisers’ existing Google Shopping and Search campaigns.

And let’s not forget the new ‘AI-organized search results.’ It’s like having a personal assistant, if your personal assistant was raised by Wikipedia and Reddit.Google says it plans to expand these pages to other categories of searches in the coming months.

We can expect legit publishers and the Forty Billion Dollar annual paid Search market to be the collateral damage of this move. Welcome to the future of search, where AI gives you what you want before you even know you want it, and then tries to sell you something you definitely don’t need.

OpenAI vs. Election Meddlers: A High-Stakes Game of Whack-a-Mole

Lastly, OpenAI is playing defense against cybercriminals trying to influence elections. It’s like a never-ending game of digital whack-a-mole, but instead of winning a stuffed animal, we get to keep our democracy. Fun!

OpenAI revealed its AI models, including ChatGPT, have been used to generate fake content aimed at influencing elections. Cybercriminals are increasingly using AI tools for malicious activities, such as creating malware and generating false content for websites and social media.

This year alone, OpenAI has neutralized over 20 attempts, including ChatGPT accounts producing articles on U.S. elections and banned accounts from Rwanda used to generate social media comments about elections. None of these efforts gained viral traction.

As the U.S. prepares for the 2024 elections, there are growing concerns about AI’s role in spreading fake or divisive information, with Homeland Security warning of potential interference from Russia, Iran, and China.

OpenAI’s ChatGPT now has 250 million weekly users, solidifying its global influence. And, yet OpenAI is stuck playing digital bouncer at Democracy’s nightclub, saying, “sorry your name’s not on the list and your bot farm invite doesn’t work here.”

---

Remember, AI won’t take your job. Someone who knows how to use AI will. Upskilling your team today, ensures success tomorrow. Customized in-person and virtual team trainings are available. Or, schedule a discovery call for customized AI consulting, including product innovation and a comprehensive strategic roadmap boost your competitive advantage with AI.

Read more: This Week in AI News: October 11, 2024