Welcome to Ad Age Datacenter Weekly, our data-obsessed newsletter for marketing and media professionals.

What DeepSeek means for marketers

“DeepSeek, a new large language model known for its low-budget computing potential, could reshape how marketers develop AI, by enabling more affordable and customizable applications for their brands,” Ad Age’s Asa Hiken reports. “But marketers are split on whether DeepSeek itself, not just its research, could be a viable part of their AI toolkits, on account of the platform’s ties to China and opacity around data collection.”

More details: “This week, DeepSeek’s surge in visibility after releasing a consumer-facing app, gave a new roadmap to ad agencies that could lead to more AI products, developed more efficiently than marketers thought was possible, according to ad leaders,” Hiken notes. “DeepSeek’s advances ... proved to be potentially as sophisticated as OpenAI, the maker of ChatGPT. It could, however, require far less computing power and money to develop, and its open-source software makes it readily available to all.”

Essential context: “Some marketers advised against using DeepSeek because of its opacity,” Hiken notes. “For instance, the Asman Group, a marketing consultancy, will not use DeepSeek because of concerns about how the platform collects data, founder and managing principal Greg Asman wrote on LinkedIn. Jon Hackett, VP of technology at agency Huge, also recommended that brands not yet migrate to the platform and away from Big Tech providers because of uncertainties surrounding DeepSeek’s training data, Hackett told Ad Age.”

Keep reading here.

Also read: An AI marketing experts tests DeepSeek

Macroeconomic news and data in a nutshell

• “Jobless claims drop to lowest level in three weeks, supporting Fed’s assessment that labor market is solid,” MarketWatch reports

• “Fed pauses interest rate cuts. What happens next may depend on Trump,” per the Los Angeles Times

• “Mortgage demand drops further, even as interest rates settle,” from CNBC

Just as you suspected, Super Bowl ads are getting less likeable

Market research and data analytics company MarketCast is gearing up to drop its take on all the commercials that air during Super Bowl LIX, but first it went back in time and crunched data around the last 12 years of Big Game ads. The analytic lens it used involves Brand Effect data, which measures if the viewing audience “remembers your advertising, likes your advertising, links your ads to your brand and is considering your brand or products because of your ads.”

Some key findings of its study, shared exclusively with Datacenter Weekly, focus on likeability:

• Simply put, Super Bowl ads have been getting less likeable over time. More specifically, MarketCast found that the most “most liked” ads aired between 2012 and 2015. “During this period, ad creative took a lot of risks—they played off gender stereotypes, were more naughty and sarcastic,” MarketCast concludes in its study.

• Looking back at that same 2012-2015 golden age of likeability, MarketCast notes that only about 30% of the top 10 “most liked” ads in each of those years featured celebrities or pro athletes.

• Across the past 12 years, “2013 saw the highest ad likeability with an average of 54%,” MarketCast notes, “followed by 2012 with 53% and a tie between 2014 and 2015 at 49%. The year with the lowest ad likeability scores was 2017 with 38%.”

• Across the past 12 years, three consumer brands—Doritos, M&M’s and Budweiser—each landed among the “most liked” six times. Only the NFL itself did better, earning “most liked” designations for its Super Bowl commercials seven times.

Check out our 2024 Super Bowl ad review here

Check out our Super Bowl archive here

Amazon helps connected TV advertisers better leverage data with new AI product

“Amazon released a new AI product in its demand-side platform focused on connected TV advertisers, helping them target brand awareness campaigns based on data that is typically applied to performance marketing,” Ad Age’s Garett Sloane reports.

The details: “On Monday, Amazon Ads announced the update to the DSP, calling the new campaign type Brand+, which works in tandem with Performance+, an AI-powered product that launched last year,” Sloane notes. “The new ad product was built to deliver video ad campaigns on Amazon and its properties, such as Twitch, as well as to third-party publishers, including Buzzfeed, Fox Corp. and Dotdash Meredith, Amazon said in its announcement.”

Essential context: “Advertisers will use machine-learning models within the DSP to target audiences that are ‘most likely to buy their products and services,’ rather than serving ads to broad demographic profiles of consumers who might not be in the market for those brands, according to Kelly MacLean, VP of Amazon DSP,” Sloane adds.

Keep reading here.

Just briefly

• “Sensitive DeepSeek data exposed to web, cyber firm says,” per Reuters

• “OpenAI, Microsoft, Trump admin claim DeepSeek trained AI off stolen data,” Mashable reports

• “Amazon accused of secretly tracking shoppers—and selling sensitive data collected through backdoor,” per the New York Post

• “Consumer loyalty in 2025—why data transparency matters more than discounts,” from Ad Age

The newsletter is brought to you by Ad Age Datacenter, the industry’s most authoritative source of competitive intel and home to the Ad Age Leading National Advertisers, the Ad Age Agency Report: World’s Biggest Agency Companies and other exclusive data-driven reports. Access or subscribe to Ad Age Datacenter at AdAge.com/Datacenter.

Ad Age Datacenter is Bradley Johnson and Joy R. Lee.