KIMP’s Picks May 29th: Weekly Highlights In Design, Marketing, AI, & More

Friday shouldn’t be about FOMO. It shouldn’t be about compulsively scrolling through feeds to catch up on the social media buzz you missed. AI news, marketing trends, social media algorithm updates – whatever you are looking for, here is a quick roundup for you. So, bid adieu to doomscrolling and start catching up on the highlights of this week. 

Welcome to KIMP’s Picks’ May 29th Edition. 

In the Marketing Realm 

Google moves Display Ads to Google Display Network in Demand Gen 

Google is moving Display Ads into Demand Gen. This helps advertisers combine the visual advertising tools in the platform under a single campaign type. Advertisers will still have access to the Google Display Network across more than 2 million sites and apps, but campaigns will now also be able to run across YouTube, Discover, Gmail, and Maps from one setup.

The rollout begins in June 2026 with a migration tool that allows advertisers to shift existing Display campaigns into Demand Gen while keeping up to 42 days of campaign learning and performance history. Google says this will reduce disruption and avoid restarting campaigns from scratch. Over time, all new Display campaigns will need to be created within Demand Gen, while older campaigns that are eligible for the transition will eventually migrate automatically.

Key marketing takeaways from Google’s announcements Google Marketing Live 2026

Google shared a post that summarizes 5 major takeaways for marketers from the major announcements that the company made at Google Marketing Live 2026. First, the focus is on the changes coming to Search as Google is placing AI at the center. 

Meanwhile, YouTube is leveraging upgraded Demand Gen features to better predict consumer behavior across Maps and feeds. E-commerce is also getting a big boost with conversational AI shopping and simplified checkout tools. On the creative side, a new Asset Studio helps teams instantly generate and test fresh, high-quality ad materials using simple text prompts. Finally, data measurements are getting an overhaul through a data command center in Google Analytics.

ChatGPT referral traffic shoots up 

New data from Similarweb shows a major spike in website traffic coming from OpenAI’s ChatGPT after the platform began displaying more prominent links to brands in its answers on May 7.

According to Similarweb, referral traffic from ChatGPT increased by nearly 150% compared to the previous week. Around 60% of those visits landed directly on brand homepages, suggesting users are increasingly clicking through AI-generated responses to explore products, services, and companies further.

The update highlights how AI search behavior is evolving beyond simple answers and becoming a real traffic source for publishers and brands. As AI assistants continue integrating clickable brand references, businesses may need to rethink SEO and content strategies to optimize for AI-driven discovery and referral traffic.

Major shifts are happening in sports marketing 

With FIFA just around the corner, now is a good time to talk about sports marketing. A recent post by Digiday discusses how sports marketing is no longer dominated by massive TV budgets and official sponsorships. 

According to Sportradar’s Nikolaus Beier, brands are now focusing on real-time fan engagement powered by live data, AI, and omnichannel technology.

Moreover, modern sports fans move constantly between platforms, devices, and formats, making traditional advertising methods less effective. So, instead of relying on static campaigns, brands are increasingly targeting emotional moments during live sports events, such as goals, momentum swings, or key match reactions, when fan attention and purchase intent peak.

This shift is significant because brands that can react in real time to fan emotions and live sporting moments will have a major advantage in attention, engagement, and conversions.

In the Content Marketing Realm 

Google says AI Search will prioritize deeper human content over basic summaries 

Google is reinforcing how content should evolve as AI-generated answers take over more simple search queries. Search Engine Land summarized what Nick Fox, SVP of Knowledge & Information at Google, said at Google Marketing Live 2026 about ranking for AI search. He spoke about how this still follows the same core rule of creating strong, high-quality content.

He stressed that AI summaries will often handle the first layer of information. This means content that performs well needs to go beyond surface explanations and answer deeper follow-up questions. In his words, the strongest content “goes one level deeper, two levels deeper” and adds real value beyond a basic overview.

This update is significant because it signals that brands and publishers will need to focus more on depth, originality, and human experience to remain visible as AI-driven search becomes the default entry point for information. 

AI Search is changing SEO from clicks to pre-click influence and brand visibility 

AI search is reshaping how users discover and evaluate information. Instead of a clear path from search to click to conversion, users now get AI-generated answers that compare options, form opinions, and influence decisions before they ever visit a website. This makes visibility inside AI answers just as important as traditional rankings.

But, the core SEO fundamentals have not changed. Google confirms that AI search still relies on the same systems that reward crawlable, indexable, and high-quality content. What is changing is how content is used. AI systems may summarise information, cite sources, or recommend brands directly, which shifts value from clicks to presence within the answer itself.

This means that a brand’s success in search will increasingly depend on how well it is understood, cited, and recommended by AI systems before users even reach the website.

In the Design Realm 

A brand identity design project that demonstrates the power of AI-assisted creativity 

A recent post from The Brand Identity explores a brand design project that involves a minimalist brand character called Vitrô designed for VitaDairy. This project reflects how AI-influenced design workflows are changing creative direction. 

Early AI-generated outputs were described as rough and unpredictable, but the team saw value in that randomness and refined it manually into a controlled design system.

Instead of forcing strict precision through prompts, designers adjusted details like facial structure, proportions, and limb length by hand. The goal was to keep Vitrô slightly abstract and emotionally flexible, allowing audiences to project their own interpretation onto the character. 

Projects like these demonstrate how brands are blending AI-driven experimentation with human creativity to build flexible, culturally resonant identities that work across physical and digital environments.

After Spotify, SoundCloud has announced a change to its app icon 

Just last week, Spotify’s decision to temporarily change its app icon to a disco-ball-inspired version in celebration of its 20th anniversary received heavy backlash. It also led to the discomorphism trend. But the brand quickly responded to criticism and announced that the original flat logo would return. 

If anything, this incident highlighted the real impact of brand designs and how much customers get invested in brands and their identities. 

Another major player in the segment, SoundCloud, has now announced some changes. But this time the changes reflect the fact that the brand is listening to its audience. In response to what its audience has been asking, the brand shared a post announcing that they are going back to the signature “orange” color that customers like. 

In the AI Realm 

Advanced music generation with Music v2 from ElevenLabs 

ElevenLabs has released Music v2, a major upgrade to its AI music model focused on improving vocals, instrumentation, and overall composition quality across genres. The update also strengthens multilingual performance, allowing lyrics and vocals to sound more natural and consistent across different languages.

The model introduces deeper creative control, including the ability to regenerate specific parts of a track without affecting the rest. It can also maintain structure across full compositions, letting users build complete songs section by section while preserving continuity. Music v2 can handle complex transitions in style, dense vocals like rap, and even integrate sound effects while keeping the music coherent.

Tailored specifically for commercial use, Music v2 is trained entirely on licensed data. This means brands can instantly generate and deploy custom audio with zero sync fees, zero clearance delays, and total peace of mind. 

Google’s massive push for AI content transparency 

As AI-generated content is expanding online, Google is rolling out a major expansion of its content transparency and verification systems. The idea is to reinforce transparency and simplify the distinction between AI-generated and real content. The update spans Search, Gemini, Chrome, Pixel devices, and Google Cloud, with an increased focus on helping users understand where digital content comes from and whether it has been altered.

At the core of this system is SynthID, Google’s watermarking technology that embeds invisible signals into AI-generated images, videos, and audio. The company says it has already been used to watermark over 100 billion images and videos and tens of thousands of years of audio content. Alongside this, Google is expanding the use of C2PA Content Credentials, an industry standard that tracks how media is created or edited, including whether AI was involved.

The company is working with partners like OpenAI, ElevenLabs, and others to broaden the adoption of watermarking standards across the industry. 

Google allows retailers to build their own virtual assistants 

AWS has introduced the Agentic Shopping Assistant (ASA) on AWS. This new AI retail solution is designed to help brands build conversational shopping assistants powered by Amazon’s generative AI expertise. The system is developed with the AWS Generative AI Innovation Center and is based on insights from Amazon’s own Alexa for Shopping technology.

The solution gives retailers a ready-made foundation that includes architecture, starter code, and expert guidance, allowing them to launch AI shopping experiences in weeks instead of years. 

Also, retailers can layer their own product data, business rules, and brand voice on top, making each assistant fully customized to their customer base. 

Amazon launches “GenAI Creators Fund”

Amazon is pushing deeper into generative AI with a new GenAI Creators Fund aimed at transforming how films, series, and digital content are produced. The initiative is led by Amazon MGM Studios and AWS, and it will fund filmmakers, creators, and startups that use AI tools across the production pipeline.

The program provides both funding and access to Amazon’s AI stack, including tools built on AWS such as Project Nara, which integrates generative models with traditional production software like Maya, Blender, and Adobe Suite. The goal is to streamline end-to-end filmmaking workflows and reduce time and cost in areas like world-building, animation, and visual effects.

According to Amazon, the technology can make large-scale storytelling more accessible, especially in animation, where production costs are traditionally high.

Google introduces AI Threat Defense 

Google Cloud has introduced AI Threat Defense, an autonomous cybersecurity platform designed to help enterprises detect, prioritize, and fix vulnerabilities at machine speed as AI-driven attacks become faster and more automated.

The company says attackers are now using AI to find and exploit security flaws in hours or days, compressing response time for security teams that still rely on manual processes. The new system aims to close that gap by combining Gemini models, Mandiant threat intelligence, Wiz risk analysis, and Google’s CodeMender remediation tools into a single automated defense framework.

A key feature is its ability to predict attack paths and generate verified code fixes inside development workflows. This reduces patching cycles from weeks to minutes in some cases. It also uses multiple AI models to improve detection accuracy across different types of vulnerabilities.

Google’s AI Overviews reportedly fail basic spelling tests 

Can Google’s AI Overviews spell “Google” correctly? Some users online say it can’t. Google’s AI-powered Search feature is under scrutiny after users spotted unusual mistakes in its AI Overviews, including incorrect answers to simple questions like counting letters in words. Reports highlighted examples where the system miscounted letters and produced incorrect spellings, raising questions about the reliability of generative AI in basic language tasks.

According to Google AI Overviews, “This happens because Large Language Models (LLMs) do not read words letter-by-letter. Instead, they process language using tokens (chunks of characters or syllables). Because the AI sees the token for “Google” as a single unit of meaning rather than individual letters, it has to guess or “hallucinate” the math when asked to break the word down into individual letters.” 

The incident follows earlier criticism of AI Overviews, which previously surfaced inaccurate or unsafe suggestions, including bizarre search results pulled from unreliable online sources during early rollout phases.

This matters because it highlights a key limitation of generative AI in search – while models are powerful at reasoning and synthesis, they can still fail at simple, structured language tasks, reinforcing the need for human verification in AI-assisted search results.

Google Search’s AI overhaul triggers user backlash as DuckDuckGo sees a surge in adoption 

Google’s major Search overhaul announced at its I/O conference is sparking backlash as the company pushes AI deeper into everyday search. The new system turns search into a conversational experience, with AI Overviews providing direct answers and a more advanced AI Mode allowing follow-up questions within results instead of just showing traditional links.

Google says these features are optional and not the default, but critics argue the shift reduces user control and introduces accuracy concerns. Some users also worry that AI summaries may replace open web results and make simple searches unnecessarily complex.

The reaction appears to be benefiting DuckDuckGo, a privacy-focused search engine that lets users opt out of AI features entirely. The company reported an 18% week-over-week rise in U.S. app installs, with even stronger growth on iOS and sustained increases in traffic to its AI-disabled search page.

DuckDuckGo says users are not rejecting AI completely, but want control over when and how it appears in search. Its platform combines optional AI tools with a default privacy-first search experience.

This highlights a growing tension between AI-first search design and rising user demand for transparency, simplicity, and control over how information is delivered. 

Social Media Feature Updates and Algorithm Changes 

Meta announces paid “Plus” subscriptions across its platforms 

There have been speculations about Meta working on paid plans across its platforms and now it’s official. The company is rolling out new paid subscription plans across Instagram, Facebook, and WhatsApp. This signals a shift toward monetizing core app features beyond advertising. The new “Plus” tier pricing starts at $3.99 per month for Instagram and Facebook, and $2.99 per month for WhatsApp.

Instagram Plus adds tools for deeper engagement and personalization, including enhanced Story insights, custom profile options, searchable viewer lists, anonymous viewing features, and expanded posting controls. 

Facebook Plus mirrors many of these engagement-focused features, while WhatsApp Plus focuses on customization in messaging, such as chat themes, pinned conversations, and personalized notifications.

View this post on Instagram

A post shared by Naomi Gleit (@naomigleit)

YouTube rolls out clearer AI content labels and wider deepfake detection tools 

YouTube is tightening its approach to AI-generated content with new tools aimed at making disclosures clearer for viewers and easier for creators to apply. Additionally, the platform is introducing visible labels on the video player for realistic AI-generated content, covering both long-form videos and Shorts.

Alongside labeling, YouTube is rolling out automatic AI detection to help creators identify when their content should be marked as AI-generated. The goal is to reduce confusion and improve consistency in how synthetic media is disclosed across the platform.

The company is also expanding access to its likeness detection technology. Over the next few weeks, all channels with users aged 18 and above will be able to use it, regardless of subscriber count, monetization status, or whether they are active creators.

Meta silently launches a Reddit-like app called Forum

Responding to the recent trends in search, where forums like Reddit are becoming reliable sources of information, Meta has silently introduced a new app called Forum. The app reshapes Facebook Groups into a more structured, discussion-driven experience similar to Reddit. The app pulls a user’s existing groups into a separate interface focused on deeper conversations, questions, and community answers.

Forum is designed around asking and answering questions, with Meta highlighting features that surface responses from across multiple groups. Users can log in using Facebook and instantly see their groups, join discussions, or discover related communities based on interests.

YouTube expands discovery and content recommendation with AI-powered Custom Feed 

YouTube is rolling out a new personalized discovery experience that lets users create their own custom content feeds using prompts. The feature allows viewers to define what they want to see, from specific interests to moods, and instantly generate a dedicated feed that updates continuously.

Users can activate the feature through a new “Your custom feed” option at the top of the Home page. By entering a prompt such as relaxing content or something outside their usual recommendations, the platform builds a tailored feed that can be pinned for easy access and adjusted at any time to refresh results.

A Final Word…

From the AI updates this week, one thing is clear – AI isn’t a good-to-have anymore. It’s becoming mainstream integration in various industries. It’s changing how platforms, tools and even entire industries are evolving. From search and ads to security, shopping, and entertainment, every major player is rebuilding around intelligence and automation.

At the same time, users are pushing back for control, transparency, and simplicity, especially in how AI shows up in everyday experiences. That tension is influencing how AI companies are building and deploying AI tools. 

For brands and creators, the opportunity is no longer just about adopting AI, but about using it in ways that actually improve clarity, trust, and relevance. Next week will likely push this even further, so the real question is who adapts fast enough without losing the human layer. So, want to stay in the loop and grab the updates before it’s too late? Stay tuned for our next KIMP’s Picks edition.