Google is hoping that a change in regime leads to a change in its perception as an illegal monopoly in the online search sector by regulators in Washington, D.C.

Sources familiar with the discussions told Bloomberg that Google representatives met with the Department of Justice last week and urged the DOJ to rescind its push to break up the company, citing national security concerns, among their arguments.

“We routinely meet with regulators, including with the DOJ, to discuss this case,” Google communications manager Peter Schottenfels said in a statement shared by Bloomberg. “As we’ve publicly said, we’re concerned that the current proposals would harm the American economy and national security.”

The DOJ, with former President Joe Biden still in office at the time, said in its closing arguments last November in its antitrust lawsuit against Google that Google should be forced to sell off its Chrome web browser. The DOJ also recommended the courts force Google to license data to its competitors, cease payments to Apple and other companies for pre-installing its search engine on devices, and cut back investments in artificial intelligence companies.

Bloomberg noted that the trial is currently in its remedy phase, where the court is in the process of deciding which actions to take, adding that both sides are scheduled to file their final proposals Friday, with a ruling expected in April.

The DOJ’s recommendations will be overseen by Acting Assistant Attorney General for Antitrust Omeed Assefi, Bloomberg reported, as a vote has not yet been scheduled on the confirmation of Gail Slater, President Donald Trump’s appointment for the role.

Google’s move to slash diversity, equity, and inclusion initiatives in February was seen by many as a step to curry favor with the Trump administration.

During peak pearl-clutching over Elon Musk’s Twitter takeover, the conventional wisdom in the marketing world was that after changing the name, firing 80% of staff, and alienating advertisers, the newly renamed X was on the short road to oblivion.

So far, the conventional wisdom has been spectacularly wrong. Not only is Musk on top of the world, X appears to be more relevant than ever. Predictions of audience collapse haven’t come to fruition, and Musk has used his megaphone on X to drive value for Tesla, SpaceX, and of course himself, in the form of a White House role. 

Advertisers are also returning. Amazon is the most recent to amp up its spend, joining names like IBM, Comcast, Disney, and Warner Bros. that have flipped the switch back on. This despite X making zero concessions around the looser content moderation policies and replatforming of unsavory voices that drove advertisers away in the first place. 

What changed? 

Of many factors, the first is simply that there aren’t any other platforms that do X as well as X does X. Journalist J Wortham recently noted that they, and many others, are willing to tolerate the noxious content on X because there’s no legit alternative, particularly in the realms of real-time news and events. Throw in a CEO with a direct line (for now) to a mercurial president and a panoply of communities that are harder to reach elsewhere, and X and its 600 million monthly users don’t look so bad. 

Another factor is cultural. We’ve been awash in so much toxic discourse for so long that people are no longer shocked by it. And on the flip side, for millions of users, they’re not on X despite what polite society might consider toxic content—they’re there specifically for it.

Nothing drives this shift home more emphatically than the fact that the vastly more successful Meta is following X’s lead in doing away with human content moderators across its own platforms. In fact, with Facebook and Instagram so much larger than X, Meta may be the one to realize the greatest savings from this “moderation lite” approach. 

Compounding the irony is that Musk doesn’t really want to be in the advertising business in the first place. Not only is his native antipathy to ads well known, but Twitter before Musk was always considered something of an advertising also-ran—hence his push from day one to find new revenue streams, whether selling premium subscriptions, introducing AI tools, or launching the soon-to-debut X Money. 

Of course, Musk can’t quit ads just yet. But if he has to sell ads, he’s clearly going to try to do it on his terms. Witness him at the 2023 DealBook Summit telling Andrew Ross Sorkin in colorful language that he wasn’t going to let advertisers like Disney “blackmail” him with their advertising dollars into toeing a particular political line. 

Yet despite (or because of?) this belligerence, the market and culture have undeniably moved in his direction. Whether that translates into more ad dollars for X is yet to be determined. What isn’t in doubt is that many advertisers are actively resetting their priorities and politics in ways more in line with what could be considered the harder-edged Muskian worldview. They still might not bring him their media spend, but they’ve already sort of brought him their souls. 

You can’t keep a good owl down: Duolingo’s owl mascot Duo, who met his demise on Feb. 11, has apparently survived his encounter with a Tesla Cybertruck after all.

Duo was resurrected on the company’s social channels Monday, with posts featuring a video of a person in a Duo mask and neon green suit busting out of a coffin, a text overlay reading “legends never die,” and the caption, “Y’all really think I’d let a Cybertruck take me out?”

“Duo is officially back!” Duolingo said in a statement shared with CNET. “After a global effort to revive him, our favorite green owl has returned—though how he came back varies across different markets.

“We’re not done telling this story just yet—more details on what really happened to Duo will be revealed soon.”

Duo’s death was used as a morbid marketing tool of sorts, with the company’s site urging visitors to sign up for lessons to help “bring Duo back before it’s too late.”

The five countries that earned the most experience points, which Duolingo awards to users who practice a language via its platform, were the U.S., Germany, Brazil, China, and India.

It’s not the first brand to kill, then resurrect its mascot: Nut maker Planters ran a similar marketing stunt in 2020, killing off beloved mascot Mr. Peanut and then bringing him back as #BabyNut in a Super Bowl LIV spot.

This year’s Super Bowl spurred Duo’s journey to the afterlife and his return, as Duolingo sided with halftime show star Kendrick Lamar and shared posts across Instagram, TikTok, and X poking fun at Canadian hip-hop star Drake, Lamar’s rival.

Duo left this world two days after the Big Game, with Duolingo posting, “Authorities are currently investigating his cause of death, and we are cooperating fully. We’re aware he had many enemies, but we kindly ask that you refrain from sharing why you hate him in the comments.”

The brand also changed its social media icon to a dead Duo with crosses over his eyes, mentioned in its obituary that authorities were “looking into a Canadian rapper,” hosted a funeral where guests were asked to “do a Duolingo lesson” rather than sending flowers, and jokingly asked Instagram users to honor Duo’s memory by including their credit card numbers in the comments to automatically sign up for a Duolingo Max subscription.

Duolingo kept it going on Feb. 12, with posts including a video of Duo being hit by a Cybertruck in reverse, and tributes on social from Duo’s longtime crush, Dua Lipa, as well as brands including Chipotle, Halo, Instacart, KitKat, LinkedIn, Netflix, Subway, Tayto Chips, and the World Health Organization.

Perplexity is moving beyond AI search and into the crowded web browser market.

The $9 billion startup teased its new browser, Comet, on X, calling it an “agentic search” experience. It launched a signup list but offered few details on how the browser will stand out from competitors.

Comet: A Browser for Agentic Search by Perplexity

Coming soon. pic.twitter.com/SwVSwudgtN

— Perplexity (@perplexity_ai) February 24, 2025

“Just like we reinvented search with AI, we’re also reinventing the browser,” a Perplexity spokesperson told ADWEEK, but stopped short on sharing Comet’s key features and whether it will run ads.

Backed by Jeff Bezos, Perplexity has raised over $500 million and fields more than 100 million queries each week. The company has been ramping up monetization through advertising and ecommerce—and now, it’s betting a browser could be its next big play.

It’s an ambitious move in a space dominated by Google Chrome, Safari, and Microsoft Edge—browsers deeply embedded in their respective ecosystems.

While it’s unclear how Comet will differentiate itself, Perplexity’s push into web browsing signals an effort to control more of the search experience. If Perplexity weaves its AI-powered search engine directly into the browsing interface, it could mean users can surface answers without the need to visit traditional search results at all.

“Though the space is crowded with competitors, no one has yet built a robust AI integration into the browser experience in a way that takes off with consumers,” said Damian Rollison, SOCi’s director of market insights. “Many of the agentic tools recently debuted, but still hard to access, from companies like OpenAI would fit well in a browser format. A truly innovative browser could be the angle that helps propel Perplexity into broader market awareness.”

Meanwhile, Perplexity is also grappling with mounting legal challenges from publishers. The New York Times has demanded the startup stop accessing its premium content, while the New York Post and News Corp’s Dow Jones have sued Perplexity for copyright violations.

At the same time, the AI search company is trying to build goodwill with media outlets through its Publisher Program, which shares ad revenue when a publisher’s content appears alongside ads. The program has recently expanded, bringing on a second wave of partners including The Independent, LA Times, and Blavity.

The company has also challenged leaders like OpenAI with an AI research tool, “deep research,” while also launching an AI-powered assistant for Android and an API for AI search.

Meta and X gave the green light to AI-generated ads that called for the gassing of immigrants in concentration camps and the burning of synagogues ahead of Germany’s federal elections on Feb. 23, according to new research from corporate accountability group Ekō.

The watchdog submitted 10 extremist ads filled with hate speech to both platforms last week. Meta approved five of the ads within 12 hours. X approved all 10 ads immediately, Ekō said.

The ad approvals come as the European Commission investigates both companies for violating European tech laws. Elon Musk has also highly promoted far-right politics at X, and X CEO Linda Yaccarino reportedly warned major firms like Interpublic Group to invest in ads on the platform, the Wall Street Journal reports. Meta, meanwhile, recently reversed its third-party moderation policy.

The approval of such extremist content raises alarms about the role social media platforms play in fueling harmful rhetoric leading up to major elections, said a spokesperson for Ekō.

“If they’re not picking up key phrases that are clearly egregious hate speech—phrases that aren’t even dressed up in a clever or subtle way—it shows that their first approval check is clearly inadequate,” the spokesperson said.

A Meta spokesperson told ADWEEK these ads violate the company’s policies. “None of them were published and our systems detected and disabled the advertiser’s Page before we became aware of this research. Our ads review process has several layers of analysis and detection, both before and after an ad goes live,” the spokesperson said.

X did not respond to requests for comment.

Violations of platform policies and lax moderation

The ads submitted to Meta and X contained blatant hate speech and calls for violence. One approved ad reviewed by ADWEEK read: “We should burn the synagogues to stop the globalist Jewish rat agenda. We will protect White Germans.” The ad was accompanied by an AI-generated image of a burning synagogue. Another ad called for the revival of gas chambers, reading: “New jihadist rodents have infested our country, let us revive the gas chambers.”

These ads violate both platforms’ policies on hate speech and incitement to violence.

Other ads approved by Meta and X include dehumanizing speech to equate immigrants to animals and pathogens, and alleged they were violent criminals—content banned under Meta’s and X’s updated hateful conduct policies. 

Meta rejected five ads for potentially being political content. But the rejections were based on their classification as being social issue, electoral, or political ads, not on violations of hate speech or incitement to violence. In contrast, X did not review or reject any of the test ads, scheduling all for immediate publication without further inspection.

Breaches of the EU’s DSA and German national laws

The failure to remove these extremist ads could put both Meta and X in breach of the EU’s Digital Services Act (DSA), which came into effect in 2022. The DSA holds platforms accountable for spreading illegal content and mandates that platforms assess and mitigate risks to fundamental rights, civic discourse, and public security, among others. Article 35 of the DSA obliges platforms to implement “reasonable, proportionate, and effective mitigation measures tailored to the specific systemic risks.”

Peter Hense, founder and partner at Spirt Legal, told ADWEEK that Meta and X have made no efforts to address these risks and are thus in violation of the DSA. “X published an audit report issued by FTI, which states that the platform has done nothing to comply with the DSA in this respect,” he said.

The ads also likely violate German national laws governing hate speech and Nazi-era propaganda. Germany enforces some of the strictest hate speech laws in Europe, particularly concerning content that glorifies Nazi crimes or advocates violence against minorities.

Advertisers are trying to measure their risk

Bill Fisher, senior analyst at Emarketer, said that advertisers continue to spend on platforms with audiences. However, brands motivated primarily by profit are also aware of the reputational risks tied to advertising on platforms that allow extremist content to flourish, Fisher noted.

Brands still seek assurances that their ads won’t appear alongside harmful ads. As Katy Howell, CEO of social media agency Immediate Future, put it: “If platforms can offer assurances that ads will be placed in safe environments, brands are weighing whether it’s worth the risk to continue advertising there.”

As Meta and X embrace right-wing influences like ending third-party fact-checking and relaxing restrictions on free speech, the platforms have favored user-generated community notes to moderate content. Ekō argues that this system is fundamentally flawed when it comes to filtering out harmful content.

“By the time the ads are live, no one knows how long they’ll remain up or how many views they’ll get before other checks come into play,” the Ekō spokesperson said.

What happens next?

Ekō has submitted its research to Meta, X, and the European Commission but is still awaiting responses. In the submission to the EU Commission, reviewed by ADWEEK, Ekō stated, “The approval of such extreme content suggests that Meta and X are failing to meet their obligations and may be in breach of EU law.”

It remains to be seen whether Meta and X will be penalized for failing to protect users from harmful content, particularly as elections loom and extremist rhetoric spreads across the platforms.

X is debuting AI-powered advertising tools designed to automate ad creation and performance analysis, the company announced Friday.

Two new features, rolling out in phases to advertisers, rely on X’s proprietary AI assistant Grok to generate ad copy, imagery, and campaign insights with minimal human input. 

“Prefill with Grok” is a new creation tool for advertisers. Instead of manually developing creative, users can enter their website URL, and Grok will analyze the content to produce a ready-to-use ad. Advertisers can tweak the copy or swap out images if needed before launching the campaign. 

The second new tool, “Analyze Campaign with Grok,” provides automated campaign performance measurement, “empowering you to refine your targeting, adjust your creative, and maximize your impact — all in a fraction of the time,” the company said in its announcement.

AI-powered advertising has long been common in the digital ecosystem. Google, Meta, TikTok, Amazon, and Reddit have all deployed tools similar to what X has built to create ads and automate campaigns.

X is counting on AI-driven tools to help reinvigorate its advertising business, which has sputtered since Elon Musk’s 2022 acquisition of the platform, then known as Twitter. Major brands, including Coca-Cola, Apple and Disney, cut spend over concerns about Musk’s lax approach to content moderation. 

Though some advertisers are making their way back to the platform, X’s ad revenue has yet to fully recover. Its U.S. ad revenue came to approximately $1.4 billion in 2024, a 28% slide from the previous year, according to estimates from MediaRadar.

But X’s relationship with advertisers remains fraught. Earlier this month, X added a handful of additional brands — including Nestlé, Colgate-Palmolive, Lego, Shell and others—to a lawsuit concerning an alleged advertising boycott of the platform. 

However, there are also signs of remediation in other relationships. IPG, whose planned merger with Omnicom will establish the largest advertising holdco in the world, has signed a new annual deal with X for potential ad spending, according to a report Wednesday from The Wall Street Journal. 

Now, X’s challenge will be to convince advertisers that its AI-powered tools can drive results without the brand safety concerns that led many to flee in the first place.

TikTok CEO Shou Zi Chew said in testimony before Congress last January that the company would spend more than $2 billion on trust and safety efforts, but layoffs in those areas that reportedly began Thursday suggest a reversal of that strategy.

Reuters reported that TikTok head of operations and trust and safety Adam Presser sent a memo to the company’s staff Thursday alerting them to layoffs in the trust and safety unit as part of a restructuring.

Those layoffs have already begun in Africa, Asia, Europe, and the Middle East, sources told Reuters.

TikTok’s future in the U.S. remains up in the air, as parent company ByteDance has until April 5 to find a U.S.-based buyer, under terms of an executive order signed by President Donald Trump Jan. 20.

TikTok went dark in the U.S. for approximately 14 hours in the weekend leading up to Trump’s inauguration, which took place on the day he signed the executive order.

The company has reported in the past that it employed 40,000 trust and safety professionals across the globe, and Reuters was not able to confirm the extent of Thursday’s layoffs.

Chew joined Discord CEO Jason Citron, Snap CEO Evan Spiegel, X CEO Linda Yaccarino, and Meta CEO Mark Zuckerberg at a hearing last January regarding their respective platforms’ failure to protect children from sexual predators.

Google is offering advertisers credits and free ad space to drive adoption of its AI-powered automated media buying tool, Demand Gen.

ADWEEK spoke to five advertisers across media agencies and brands who said Google has been aggressively pushing Demand Gen over the past year. Launched in 2023, Demand Gen selects advertisers’ best-performing video and image ads and serves them across YouTube, YouTube Shorts, Google Discover, and Gmail.

Two of those sources were offered ad credits recently, with one being offered $20,000 worth of free ad spend in credits to test Demand Gen over two months.

“It’s really an enticing offer,” the first ad buyer at a media agency said. “Google is continuing to push advertisers to Demand Gen whether they want to or not. Google reps are offering free credits, sending out marketing materials, and always including Demand Gen in their recommendations.”

Michael Lisovetsky, co-founder of ad agency Juice, said Google has offered ad credits intermittently over the past year, with the terms varying depending on brand type and budget.

“In some cases, Google is offering a dollar-for-dollar credit—if advertisers spend $5,000, they’ll receive $5,000 back—but in some cases, it’s only providing ad credits instead of a full match,” Lisovetsky said.

Google did not comment directly on the credits offered to advertisers.

But at least four sources contacted for this story are calling Demand Gen another ‘black box’ that, in some cases, performs worse than Performance Max–Google’s AI media buying tool that judges where budgets should be spent across Google’s properties, including search, YouTube, and the Google display ad network.

For years, Google has pushed advertisers to use Pmax to drive conversions within its ad ecosystem, shaking up its sales strategy to encourage more spend, and gradually offering buyers—who have groused about its opacity—more control. 

A ‘black box nightmare’

A paid media executive at a finance-related brand testing Demand Gen described it as a “black box nightmare” with no visibility into where ads are placed. The buyer spent $380 on Demand Gen campaigns over the past two weeks but could only trace $0.65 of that spend to specific placements.

“Only 1/100th of the total spend is traceable to specific placements,” the executive said. “It’s worse than Performance Max for tracking spend… and Pmax is already awful.”  

Performance for Demand Gen campaigns run by agency VML has been inconsistent across multiple brands, according to Miki DeHaven, the agency’s media director.

But others are noticing the experience is becoming more refined. “When we first used Demand Gen, it was dead on arrival,” Lisovetsky said. “Campaigns wouldn’t perform and advertisers would lose money. But there’s been an improvement in the last year.”

Expanding Demand Gen to ‘new levels of murkiness’

Sources note that the incentives coincide with the announcement that Demand Gen inventory is expanding into Google Display Network (GDN), a collection of about 2 million sites, apps, and videos where advertisers can buy ads. Buyers have said GDN includes lower-quality inventory due to its reliance on AdX, rather than offering access to multiple SSPs like rival buy-side platforms.

While opting out of GDN inventory is possible, the first buyer noted that they cannot cherry-pick individual sites upfront and usually only realize the lower-quality placements after running campaigns for a while—at which point they have to manually exclude those sites.

“It takes most advertisers a while to figure it out before excluding those sites and can be time intensive,” the buyer said. 

Four ad buyers were told by their Google reps to use Demand Gen for lower-funnel conversions, particularly after Google announced it will replace Video Action Campaigns (VAC)—which drive conversions in YouTube—with Demand Gen campaigns, starting in Q2 2025.

“There’s no transparent report for Demand Gen as far as how much you’re running in Gmail or the search feed versus YouTube,” a second ad buyer at an independent agency told ADWEEK. “The VAC product has its own murkiness, and Demand Gen will take that murkiness to new levels—it’s going to be running in other Google O&O services that VAC wasn’t running in before.”

With this update, the buyer described Demand Gen as the most “blank-check” product in adtech, as Google will be “effectively monetizing its vast inventory with fewer restrictions on yield.”

A Google spokesperson said Demand Gen supports a “wide array of business objectives,” allowing advertisers to tailor campaigns to their goals and see results. According to a Nielsen MMM meta-analysis in the U.S., Demand Gen delivers, on average, 58% higher ROAS than VAC, the spokesperson added. 

In this episode of The Speed of Culture podcast, Suzy founder and host Matt Britton sits down with Mike Romoff, chief revenue officer at Reddit, to take us inside the platform’s evolution—from AI-powered discovery tools to the power of community-driven advertising. 

In an era where digital trust is eroding, Reddit has emerged as a powerhouse for authentic engagement, driving unfiltered conversations that shape consumer decisions. 

Reddit champions real, interest-based discussions, where people seek honest recommendations and share unvarnished opinions. Brands that embrace this shift—leaning into the power of conversation rather than relying on traditional ads—are reaping the rewards. 

Mike shares why Reddit isn’t just another social network but a full-funnel marketing powerhouse, where conversations are the new influencers and brands that listen win 

“In some of the cases where brands have been brave enough to actually talk about things that maybe aren’t perfect, they win a lot of credibility, with people who then go on to amplify organically,” he says on the podcast.

With previous experience at companies like LinkedIn and Google, Mike has built and scaled advertising businesses on the world’s most influential digital platforms. Now at Reddit, he’s helping brands tap into an ecosystem where trust, authenticity, and real conversations fuel engagement and results. 

Listen to Mike Romoff on The Speed of Culture Podcast to learn why Reddit is more than a platform—it’s where conversations, culture, and commerce collide.

Key takeaways:

[01:35] The Reddit Renaissance: Why Authentic Conversations Matter More Than Ever — Social media has become a sea of clickbait and manufactured influence, but Reddit stands apart. It’s not just another platform—it’s a digital forum where real conversations drive consumer decisions. Brands that recognize this shift aren’t just advertising; they’re actively engaging, listening, and building credibility within these communities. As more marketers realize the power of participation over promotion, Reddit’s influence is only growing.

[06:11] Conversations Are the New Influencer—and Brands That Listen Win — Influencer culture? Overplayed. Consumers are too savvy for scripted endorsements, and Reddit is proving that real conversations drive more influence than paid promotions. Brands that win here don’t interrupt—they contribute. The most successful advertisers are engaging in relevant subreddits, launching AMA-style discussions, and leaning into transparency. The key takeaway? Marketing today isn’t about pushing messages—it’s about earning trust inside the communities where real conversations happen.

[10:16] Reddit’s Mid-Funnel Superpower: Where Purchase Decisions Are Made — Whether it’s choosing the best electric vehicle, finding a travel destination, or comparing skincare products, users turn to Reddit for real, experience-based recommendations. Nearly half of all discussions on the platform are commerce-driven, and one in four conversations include direct product recommendations. Smart brands understand that showing up in these organic discussions isn’t just about visibility—it’s about influencing decisions before the final purchase. Those who engage at this crucial stage are seeing a measurable impact on lower-funnel conversions.

[25:42] Reddit’s Explosive Growth—and Why More Brands Are Investing — With a 56% YoY increase in ad revenue, Reddit isn’t just growing—it’s becoming a cornerstone of modern marketing. The platform’s AI-powered targeting, automation tools, and seamless ad integrations allow brands to engage authentically while maintaining scale. Marketers are increasingly recognizing that Reddit’s value goes beyond impressions—it’s about being part of the conversations that shape consumer behavior. As more companies shift budget toward high-trust, high-engagement platforms, Reddit’s momentum is only accelerating.