AI Briefing: Generative AI investments expected to rise alongside greater risks

Think the generative AI boom is bound to bust? A new report from the International Data Corporation says it’s time to buckle up.

Last week, the market intelligence firm released a new forecast predicting that generative AI investments will rise from $16 billion this year to a staggering $143 billion in 2027. While generative AI accounts for just 9% of overall AI spending in 2023, the firm expects that will increase to 28% within five years. IDC also noted generative AI’s growth rate is more double the growth rate of overall AI spending, and growing at a compound annual rate nearly 13 times higher than worldwide IT spending in the same time frame.

“When it comes to looking at AI strategy or even a data strategy in general, I think of generative AI kind of like this iceberg,” said Khendr’a Reid, a principal data strategy technical specialist at Amazon Web Services. “You see the tip is to generate AI, but it’s everything underneath it.” (Reid said this during one of the dozens of AI-related talks at Advertising Week in New York last week.)

Despite the rapid growth, the evolution of AI could also be challenged by increased data privacy protections in Europe and the U.S. Several more states are expected to enact newly passed privacy laws this year and others will take effect in 2024 and 2025. Earlier this month, California Governor Gavin Newsom signed “The Delete Act,” a new law that will enact new regulations for data brokers in the state.

Brian Quinn, President and GM at AppsFlyer, thinks the two landscapes “are really crashing in and these two forces could be at odds if not carefully managed.” This creates a “fundamental challenge” and “one of the largest challenges for marketers moving forward.”

“Privacy laws and restrictions are all around the principle of data minimization and protection,” Quinn said during another talk at Advertising Week. “And so a lot of what marketers have been contending with is [how to] get the same output and deliver the same experiences while protecting my consumer privacy with the loss of signal.”

Here’s a quick look at some of the other reports released last week that highlight potential opportunities and risks related to AI:

  • In a recent survey of 140 U.S. marketers conducted by research firm Advertiser Perceptions and and data management company Claravine, about half said they wouldn’t know if an ad ran next to unsuitable content until a day later or longer.
  • The growth of AI could create new cybersecurity concerns for brands. In a new report from the compliance services provider CSC, 43% of Forbes Global 2000 companies don’t have control over their branded “.AI” domain names and 49% of .AI brand domains remain unregistered — potentially exposing them to fraud, brand infringement and other risks. CSC also noted domain-disputed cases involving .AI extensions increased 350% in 2023 as companies discovered .AI domains were “misappropriated” by third parties.
  • Other reports highlight some of the potential benefits of AI. In Twilio’s survey of 2,450 business leaders, 54% plan to spend more on “AI-powered campaigns” next year, while 38% plan to use chatbots in their marketing efforts and 28% expressed concerns about data privacy.
  • A separate survey earlier this month by Deloitte Digital found early generative AI adopters had a 12% return on their investments, but the report didn’t detail what that investment was. The survey also found that 26% of marketers are already using generative AI while another 45% plant to use it by the end of 2024.

Prompted Products and other AI news

  • Universal Music Group and other music publishers filed a lawsuit against Anthropic, creator of the chatbot Claude. In the lawsuit filed last week in Tennessee, the music companies allege the Amazon-backed AI company violated copyright laws by distributing music lyrics in its large language model.
  • Jasper AI has added a new AI ‘copilot” for marketing teams, which will help with personalized writing, analytics and other tools.
  • Microsoft plans to release its new Copilot on Nov. 1, which will cost $30 a month and help users with a range of tasks including across various Microsoft 365 products. 
  • Skylum has added new AI photo-editing abilities to its Luminar Neo software, echoing moves by Adobe and others in recent months.

Other stories from Digiday and beyond

  • “OpenAI’s Sam Altman has ‘deep misgivings’ about people becoming friends with AI” (Business Insider)
  • “We don’t want to set up global AI regulator, says UK tech secretary” (Politico)
  • “NYC Releases Plan to Embrace AI, and Regulate It” (Bloomberg)
  • Marketers, agency execs were abuzz about generative AI, brand IP (Digiday)
  • Gen Z workers speak out on AI ethics, data security, career disruption (Digiday)
  • Advertising Week Briefing: Why content creators are hopeful that AI will not replace them (Digiday)

AI Hot Takes

Here are some of the things people said about AI during Advertising Week New York 2023: 

  • “Yesterday I watched Star Wars in the style of Wes Anderson on social and that was really cool, but I’ll bet neither of those parties got paid for that piece of content. And as creatives make up a large portion of my loved ones, I’d definitely like to see some sort of structure around making sure that the people who are creating the content that becomes seed data are being compensated in some way, ideally appropriately. But just in general at all would be great.” — Marykate Byrnes, director of Media & Growth at La Colombe
  • “The generative models, they’re very large, they’re very fun, they’re helpful, but there’s a lot of variation in there. So there’s still more tuning to be done.” — Jen Wong, COO of Reddit 
  • “I think there has to be regulation. The debate now is who regulates it, how it is regulated and who punishes those who fall afoul of the regulations. Because I don’t think you can regulate it at a single governmental level, I don’t think you can self-regulate.” — Alex Hesz, chief strategy officer of Omnicom
  • “Once you use anything, everyone’s gonna know about it and you can’t get it back… There are rules. There are certain industries, like banking, for example, it’s a hard no. It’s not something that they wanna be getting into at the moment.” (CQ) — Digitas NA CEO Amy Lanzi
https://digiday.com/?p=522892

More in Media Buying

One media agency gets vocal against principal media — and hopes others will too

Acadia’s Jared Belsky is fomenting awareness of — and resistance to — principal-based media on the part of agencies for their clients. And he’s going public with his efforts starting today.

3 people on different sides of politics sitting at a table. Tension doing political ad research

How AI shaped the 2024 election: From ad strategy to voter sentiment analysis

With Election Day here, it’s worth taking a look at some of the ways ad-tech firms, political startups and political agencies have used LLMs and machine learning.

Presidential ads mark a divide as both candidates spend big in battleground states

Election Day is here, and new ad spending data reveals the intense and targeted push seen in swing states as campaigns for Vice President Kamala Harris and former President Donald Trump delivered last-minute messaging.