AI Search
Catch up on select AI news and developments from the past week or so:
Google signals future advertising opportunities within Gemini AI interface. Google has left open the possibility of introducing advertising into its Gemini AI assistant, signaling a major potential shift in digital advertising. As AI-generated answers reduce traditional search traffic, integrating ads into conversational interfaces could create new revenue streams. The move would require new ad formats embedded within AI responses and raises questions about user trust, transparency, and regulatory compliance. Industry observers expect experimentation to begin soon as Google balances monetization with user experience.
Importance for marketers: AI interfaces may become the next major advertising channel, requiring new strategies for integrating brand messaging into conversational experiences.
Shopify prepares for agentic shopping as AI becomes the new front door to commerce. Shopify is investing heavily in agent-driven commerce, where AI systems act as personal shoppers that discover, compare, and purchase products on behalf of users. These agents aim to provide deeper personalization than traditional search by learning user preferences and surfacing relevant products more effectively. Shopify is developing tools such as Sidekick and new protocols to support agent interactions with merchant data. The shift could expand online retail and improve product discovery, especially for smaller brands.
Importance for marketers: Discovery is shifting from search engines to AI agents. Brands will need to optimize for agent-driven recommendations rather than traditional search rankings or marketplace visibility.
Study finds most content retrieved by ChatGPT is never cited in answers. New research shows that only 15% of webpages retrieved by ChatGPT are ultimately cited in final responses, highlighting a major shift in how visibility works in AI-generated answers. The analysis reveals that citation selection depends on relevance within synthesized responses, not just retrieval or ranking. Additional internal queries expand the pool of potential sources, further complicating optimization strategies. High search rankings still correlate with citations but do not guarantee inclusion.
Importance for marketers: Visibility in AI answers requires more than ranking well. Content must align closely with how AI systems synthesize and select information, reshaping SEO into a new optimization challenge.
LinkedIn rebuilds feed ranking with LLMs and transformer-based recommender systems. LinkedIn has overhauled its feed recommendation system using large language models and transformer architectures to unify ranking and retrieval. The new system improves content relevance by analyzing semantic meaning and user behavior patterns, while enabling faster processing and broader content discovery beyond immediate networks. With the feed serving as a primary channel for both organic and paid content, the redesign has significant implications for how visibility is determined across the platform.
Importance for marketers: Content visibility on LinkedIn will increasingly depend on AI-driven relevance signals, requiring more sophisticated content strategies aligned with user intent and engagement patterns.
Visa prepares infrastructure for AI agents to initiate payments. Visa is testing systems that allow AI agents to initiate transactions on behalf of users, shifting the traditional model of human-driven payments. The initiative focuses on authentication, consent, and compliance as software agents gain the ability to make purchasing decisions within defined rules. Early pilots explore use cases such as automated procurement and recurring purchases, while addressing fraud, audit, and regulatory concerns. This signals a broader move toward agent-driven commerce where software, not people, executes transactions.
Importance for marketers: This introduces a future where AI agents, not consumers, make buying decisions. Marketing strategies will need to influence algorithms and decision frameworks, not just human preferences.
NVIDIA launches open platform for building enterprise AI agents. NVIDIA introduced its Agent Toolkit, an open platform for developing autonomous AI agents capable of reasoning, acting, and completing complex enterprise tasks. The toolkit includes OpenShell for secure runtime environments, Nemotron models, and AI-Q agent blueprints that combine open and frontier models to reduce costs while maintaining high accuracy. Major enterprise software providers are integrating the platform to power agent-driven workflows across industries. The initiative positions AI agents as a foundational layer in enterprise software.
Importance for marketers: This accelerates the shift toward agent-driven business processes. Marketing operations, analytics, and customer engagement will increasingly be handled by autonomous systems rather than manual workflows.
Alibaba launches Wukong agent platform as competition intensifies in enterprise AI. Alibaba has introduced Wukong, an enterprise AI platform designed to manage multiple agents performing tasks such as document editing, approvals, and research. The system integrates with messaging platforms and enterprise tools, reflecting a shift toward agent-based workflows in business environments. The launch comes amid internal restructuring and increased competition in AI, as companies race to develop platforms that support autonomous task execution across organizations.
Importance for marketers: Enterprise adoption of AI agents will transform workflows across departments, including marketing, enabling more automation in content creation, analysis, and campaign execution.
Picsart launches AI agent marketplace for creators to automate content workflows. Picsart has introduced an AI agent marketplace that lets creators deploy specialized assistants to handle tasks like resizing content, remixing visuals, editing product images, and optimizing online stores. Agents can analyze trends, recommend improvements, and execute tasks with configurable autonomy levels. Some agents integrate with platforms like Shopify and messaging apps, enabling asynchronous work and ongoing optimization. The rollout reflects growing demand for agentic tools that shift creators from manual execution to high-level direction and oversight.
Importance for marketers: This moves creative production toward agent-managed workflows. Marketers will increasingly direct systems rather than execute tasks, changing how campaigns are built, optimized, and scaled.
Meta-backed Manus brings AI agents onto personal devices with desktop app. Manus has launched a desktop application that allows its AI agent to operate directly on users’ local devices, interacting with files, applications, and workflows. The move expands agent capabilities beyond cloud-based environments, enabling tasks like file organization, coding, and app control. While the system includes safeguards requiring user approval, it raises new security and privacy considerations. The release aligns with broader industry momentum toward locally deployed, highly autonomous AI agents.
Importance for marketers: Local, device-level agents could transform productivity by automating workflows across tools, enabling marketers to manage campaigns, assets, and data more seamlessly through AI assistants.
The Trade Desk tests AI-driven campaign creation using Anthropic’s Claude. The Trade Desk is running a closed beta that allows advertisers to create programmatic campaigns using a large language model interface powered by Claude. The approach positions AI as the entry point for campaign setup, raising questions about defaults, transparency, and how optimization decisions are made and explained. The move reflects broader industry experimentation with AI-driven campaign creation across major ad platforms.
Importance for marketers: Campaign creation is shifting from manual configuration to AI-driven interfaces, potentially reducing the need for specialized expertise while raising new concerns about control and transparency.
Facebook Marketplace adds AI-generated replies and automated listings for sellers. Facebook Marketplace is introducing AI features that automate responses to buyer inquiries and assist with listing creation. Sellers can use AI to draft replies, generate product descriptions, and suggest pricing based on similar items. The tools aim to reduce manual effort and streamline interactions, while additional features provide insights into seller profiles and activity. These updates expand the role of AI in peer-to-peer commerce and transaction management.
Importance for marketers: AI-driven automation in commerce platforms will change how brands and sellers engage with customers, making responsiveness and efficiency increasingly algorithm-driven.
Google expands personalized Gemini AI to all US users. Google is rolling out its Personal Intelligence feature to all US users, allowing Gemini to draw on data from connected apps like Gmail, Photos, and YouTube to deliver more context-aware responses. Previously limited to paid tiers, the feature now reaches free users and works across Search, Chrome, and the Gemini app. It remains opt-in, with controls for disconnecting data sources. The system uses contextual signals to tailor recommendations and assistance without directly training on private content.
Importance for marketers: This deepens personalization at scale. Marketing strategies will need to account for AI systems that shape recommendations based on individual user data rather than generic targeting.
AI-powered advertising surges as major driver of US ad growth. AI-driven advertising is projected to grow 63% in 2026, reaching $57 billion and accounting for a significant share of total ad spend. Platforms that automate targeting, bidding, and optimization are gaining adoption across both small and large advertisers. Despite concerns about transparency and control, many brands are prioritizing performance and efficiency, increasingly trusting automated systems to manage campaigns. Growth is expected to continue at a strong pace through the end of the decade.
Importance for marketers: This confirms that AI-driven campaign automation is becoming the dominant model. Marketers will need to adapt to systems where performance is optimized by algorithms rather than manual control.
Teneo and Thoughtworks launch AI venture to help executives operationalize AI. Teneo and Thoughtworks have formed a joint venture aimed at helping enterprise leaders translate AI ambition into practical applications. The initiative combines executive advisory capabilities with engineering expertise to build custom AI tools for areas such as product development, investor relations, and regulatory strategy. The partnership reflects a growing demand for guidance at the leadership level, as organizations struggle to bridge the gap between strategy and implementation.
Importance for marketers: Successful AI adoption will increasingly depend on aligning strategy with execution, requiring marketing leaders to demonstrate measurable business outcomes from AI investments.
Anthropic scales global marketing with a single AI-augmented operator. Anthropic operated much of its global marketing with a single growth marketer supported by internal AI tools. Using its Claude Code system, the marketer automated ad creation, campaign execution, and analytics, reducing tasks that once took minutes or hours to seconds. This approach enabled rapid experimentation and high output without a traditional team structure. The company continues to invest in brand campaigns while maintaining a lean internal operation, reflecting a broader shift toward AI-augmented roles replacing larger functional teams.
Importance for marketers: This is a clear signal that AI can compress entire marketing functions into smaller, more technical roles. Marketers who can operate AI tools effectively will outperform larger teams that rely on traditional workflows.
Adobe launches custom AI image models trained on brand-specific assets. Adobe has released Firefly Custom Models in public beta, allowing users to train AI image generators on their own creative assets. These models preserve brand-specific elements such as style, color, and character consistency across outputs, enabling scalable content production without losing visual identity. The models are private by default, and Adobe includes safeguards to ensure users have rights to training data. The tool integrates into existing workflows, making it easier to generate large volumes of on-brand creative assets efficiently.
Importance for marketers: This solves one of the biggest challenges in generative AI: brand consistency. Teams can now scale content production while maintaining visual identity, making AI more viable for enterprise brand work.
Gamma expands into AI-generated marketing assets to compete with Canva and Adobe. Gamma has launched an image-generation tool designed to create marketing assets such as social graphics, charts, infographics, and presentations from text prompts. The platform combines templates with AI tools and integrates with a range of productivity and automation platforms to support data-driven content creation. Positioned between professional design tools and legacy presentation software, Gamma aims to serve business users who need visual communication without specialized design skills, while continuing rapid growth in users and revenue.
Importance for marketers: This further lowers the barrier to producing high-quality visual content, enabling more teams to create campaigns without dedicated design resources and accelerating content velocity.
Microsoft restructures Copilot and doubles down on building frontier AI models. Microsoft is reorganizing its AI efforts by merging its commercial and consumer Copilot teams and shifting leadership focus toward developing in-house frontier models. The move aims to create a more unified product experience while reducing dependence on external partners. Leadership emphasized the importance of integrating AI models, applications, and workflows into a cohesive system. The changes reflect intensifying competition among major tech firms to control both the underlying models and the user-facing platforms.
Importance for marketers: Platform consolidation will shape how marketers access and use AI tools, making it increasingly important to understand and adapt to integrated ecosystems rather than standalone solutions.
OpenAI plans unified desktop superapp combining ChatGPT, Codex, and browser tools. OpenAI is consolidating its products into a single desktop superapp that integrates ChatGPT, its Codex coding assistant, and the Atlas browser. The move aims to reduce fragmentation and focus on core experiences as competition intensifies. Leadership has emphasized prioritizing successful products and eliminating distractions, signaling a strategic shift toward tighter integration and clearer product direction. The mobile app will remain unchanged, while the desktop experience becomes the central hub.
Importance for marketers: This points to a future where AI platforms become primary work environments. Marketers will increasingly operate within unified AI ecosystems that combine research, creation, and execution.
OpenAI may integrate Sora video generation directly into ChatGPT. OpenAI is reportedly planning to bring its Sora video-generation capabilities into ChatGPT, making advanced video creation more accessible within its primary interface. The integration would streamline workflows and increase adoption but also raises concerns about misuse, including deepfakes and copyright issues. As competition intensifies, the move reflects a broader trend toward consolidating capabilities into unified AI platforms.
Importance for marketers: Easier access to video generation will accelerate content production while increasing the need for governance around authenticity, brand safety, and responsible use.
Mistral launches Forge to help enterprises build custom AI models from scratch. Mistral has introduced Forge, a platform that allows organizations to train AI models on their own data rather than relying on prebuilt systems. Unlike typical approaches that layer data onto existing models, Forge enables full customization, giving companies greater control over performance, behavior, and compliance. The platform includes tools, infrastructure, and embedded engineering support to help enterprises develop and deploy tailored AI systems. Early adopters include major industrial and government organizations.
Importance for marketers: Custom-trained models will allow companies to build AI systems aligned with their brand, data, and customers, creating competitive advantages in personalization and proprietary insights.
Mistral releases Small 4 open-source model with unified capabilities and improved efficiency. Mistral has released Small 4, an open-source model under Apache 2.0 that combines reasoning, instruction-following, and multimodal capabilities into a single system. Built with a mixture-of-experts architecture, it delivers faster performance, lower latency, and higher throughput than its predecessor while maintaining strong benchmark results. The model supports large context windows, configurable reasoning depth, and deployment across cloud and on-prem environments, making it suitable for enterprise customization and large-scale applications.
Importance for marketers: More capable open models lower costs and increase flexibility, enabling companies to build custom AI systems without relying solely on proprietary platforms.
Trustpilot positions review data for AI-driven commerce as search behavior shifts. Trustpilot is expanding partnerships with e-commerce platforms as AI-driven shopping changes how consumers discover products. The company sees its dataset of user reviews as increasingly valuable to AI systems that recommend purchases. Traffic from AI-based search is rising sharply, while traditional search patterns shift toward conversational interfaces and embedded commerce experiences. Partnerships across major platforms suggest a growing ecosystem where transactions occur within AI environments rather than on retailer websites.
Importance for marketers: Reputation data and third-party signals will play a larger role in AI-driven discovery, making reviews and trust signals critical inputs for visibility in agent-based commerce.
Study finds AI summaries increase purchase intent despite high error rates. Research shows that consumers are significantly more likely to buy products after reading AI-generated summaries of reviews, even though those summaries contain high rates of inaccuracies and hallucinations. Participants exposed to AI summaries expressed purchase intent far more frequently than those reading original reviews, highlighting how framing and presentation influence decision-making. The findings suggest that AI can distort perception by emphasizing certain details and introducing subtle biases, even when the underlying information is flawed.
Importance for marketers: AI-generated content can strongly influence purchasing behavior, creating both opportunity and risk for brands in how information is presented and interpreted.
US bets on robotics, infrastructure, and market-led innovation to win AI race. US policymakers and industry leaders are shaping long-term AI competitiveness through three key areas: robotics, infrastructure, and reduced regulation. While the US leads in foundational AI models, China dominates industrial robotics deployment, raising concerns about future competitiveness. At the same time, investment in data centers is accelerating to support AI growth, though local resistance is increasing. The government is shifting toward a market-driven innovation model, with private sector funding now dominating research and development. This approach places significant influence in the hands of large technology firms and venture capital.
Importance for marketers: The scale and direction of AI infrastructure investment will shape the capabilities, costs, and platforms marketers rely on for years to come.
Debate intensifies over AI job disruption after OpenAI cofounder removes analysis. A widely shared analysis ranking jobs by exposure to AI was taken down after being misinterpreted, highlighting ongoing uncertainty around AI’s labor impact. The analysis suggested that digital and white-collar roles such as accounting, customer service, and software development face higher exposure, while hands-on jobs may be less affected. Separate research points to similar patterns but emphasizes that real-world impact depends on demand, adoption rates, and economic factors. Meanwhile, layoffs attributed to AI continue, though critics argue other business factors are also driving workforce reductions.
Importance for marketers: This reinforces that marketing roles, especially digital and analytical ones, are highly exposed to automation. Teams will need to rethink skills, workflows, and value creation as AI increasingly handles execution-level tasks.
Encyclopedia Britannica sues OpenAI over alleged misuse of training data. Encyclopedia Britannica has filed a lawsuit alleging that OpenAI used its proprietary content without permission to train AI models, resulting in outputs that replicate or closely resemble its materials. The case adds to a growing number of legal challenges from content owners seeking compensation or restrictions on AI training practices. OpenAI maintains that its methods fall under fair use, while plaintiffs argue that AI systems are diverting traffic and value from original sources.
Importance for marketers: Ongoing legal battles over training data will influence how AI tools are built and used, affecting content sourcing, licensing, and brand risk considerations.
ByteDance delays global rollout of video AI model amid copyright disputes. ByteDance has paused the global launch of its video-generation model following copyright concerns raised by major entertainment companies. The system, designed to create cinematic content from prompts, had drawn attention for its capabilities but also for generating content resembling protected intellectual property. Legal challenges have forced the company to implement safeguards and reassess its release strategy. The situation underscores the ongoing tension between rapid AI innovation and the need to address intellectual property rights.
Importance for marketers: Copyright risks will shape how AI-generated video and creative assets are used in campaigns, requiring careful attention to content ownership and compliance.
Mystery AI model revealed as Xiaomi system built for agent workflows. A previously anonymous high-performance AI model circulating on developer platforms has been identified as an internal Xiaomi system designed to power AI agents. The model, featuring a massive context window and strong reasoning capabilities, sparked speculation it was a next-generation release from another major lab. Its rapid adoption highlights growing interest in agent-oriented models that can handle complex, multi-step tasks. The release also reflects the competitive pace of AI development and the increasing role of stealth testing in model deployment.
Importance for marketers: Advances in agent-focused models will enable more autonomous marketing systems capable of executing campaigns, analysis, and optimization with minimal human input.
Cursor releases Composer 2 with improved coding performance and lower costs. Cursor has launched Composer 2, a new in-house coding model optimized for long-horizon, agentic development tasks. The model delivers significant performance gains over its predecessor while reducing costs by up to 86%. It is tightly integrated into Cursor’s environment, enabling multi-step workflows such as editing files, running commands, and iterating across projects. While it does not outperform top models across all benchmarks, it offers a strong cost-to-performance balance and emphasizes practical usability within its platform.
Importance for marketers: Continued improvements in coding agents will further automate web development, experimentation, and martech integration, reducing reliance on engineering resources for marketing execution.
Microsoft releases MAI-Image-2 with improved realism and design capabilities. Microsoft has introduced MAI-Image-2, a new text-to-image model developed by its superintelligence team. The model improves significantly over its predecessor, delivering more realistic images, better lighting, and stronger text rendering for practical use cases such as posters and infographics. It ranks near the top of current benchmarks but still trails leading models from competitors. The tool is being integrated into Microsoft’s products and will expand through API access for developers.
Importance for marketers: Improvements in image generation quality and text rendering make AI-generated visuals more usable for real campaigns, reducing the need for traditional design production in many scenarios.
OpenAI brings GPT-5.4 mini to free ChatGPT users with faster performance. OpenAI has released GPT-5.4 mini and nano models, with the mini version now available to free ChatGPT users. The models offer improved performance in coding, reasoning, multimodal tasks, and tool use while delivering faster responses and lower costs. Designed for latency-sensitive applications, they support workflows such as code editing, debugging, and real-time analysis. The rollout reflects a broader trend toward smaller, more efficient models that balance capability with speed and scalability.
Importance for marketers: More powerful AI capabilities becoming available at lower cost and wider access will accelerate adoption, enabling more teams to integrate AI into everyday marketing workflows.
Nvidia CEO predicts massive AI demand and highlights shift toward inference computing. Nvidia CEO Jensen Huang outlined a vision of continued rapid growth in AI demand, predicting a massive backlog for AI chips and emphasizing the transition from training to inference computing. He described AI as a foundational platform shift comparable to past computing revolutions and highlighted Nvidia’s strategy to maintain leadership through both hardware and new market segments. Despite growing competition and regulatory challenges, the company expects sustained expansion driven by widespread AI adoption.
Importance for marketers: Continued investment in AI infrastructure signals long-term expansion of AI capabilities, reinforcing that AI-driven marketing tools and platforms will keep advancing rapidly.
UK considers labeling AI-generated content as part of copyright reforms. The UK government is exploring new regulations that could require labeling of AI-generated content, alongside broader efforts to address copyright, consent, and compensation for creators. The proposals aim to balance innovation with protections against misuse, including unauthorized digital replicas and unclear ownership of AI-generated works. Policymakers have not settled on a final approach, signaling ongoing uncertainty about how to regulate AI training and output. The initiative reflects increasing global pressure to establish clearer legal frameworks for AI content.
Importance for marketers: Potential labeling requirements could reshape content strategies, requiring transparency about AI use and influencing how audiences perceive and trust branded content.
xAI undergoes major rebuild as competition intensifies in AI coding and agents. Elon Musk’s xAI is restructuring after internal challenges and leadership departures, as it struggles to compete with rival coding tools and AI platforms. The company is rebuilding its systems and teams while pursuing ambitious projects, including agents capable of performing complex digital tasks. Efforts to integrate AI with robotics through initiatives involving Tesla highlight a broader vision, though execution has faced setbacks. With increasing competition from established players, xAI faces pressure to demonstrate progress and deliver commercially viable products.
Importance for marketers: Competitive pressure among AI vendors will accelerate innovation, giving marketers access to more advanced tools but also increasing fragmentation across platforms.
Anthropic tests Claude Dispatch for remote control of desktop AI workflows. Anthropic has introduced Dispatch, a feature that allows users to control a desktop AI workspace remotely from a mobile device. The tool connects a mobile app to a Mac-based AI session, enabling tasks such as searching files, summarizing content, and interacting with connected apps. While still experimental and inconsistent, it demonstrates progress toward seamless cross-device AI orchestration. The feature builds on existing capabilities that allow AI systems to interact with local environments and workflows.
Importance for marketers: Cross-device AI control points to a future where marketers can manage workflows, content, and analysis from anywhere, increasing flexibility and responsiveness in campaign execution.
CFO pressure, not CMO leadership, will drive AI adoption in agencies. Industry veteran Martin Sorrell argues that large-scale AI adoption in marketing organizations will be driven primarily by financial pressure rather than marketing leadership. He points to rising production costs and the ability of AI to generate massive volumes of creative assets as key factors forcing change. The shift reflects structural challenges in the advertising industry, where digital platforms dominate spend and efficiency gains are becoming critical.
Importance for marketers: AI adoption will be driven by cost efficiency and performance demands, not just innovation, forcing marketing teams to justify their value in increasingly automated environments.
Google introduces Stitch as an AI-native design canvas enabling ‘vibe design’. Google is evolving Stitch into an AI-native software design environment that turns natural language into high-fidelity UI and interactive prototypes. The platform introduces an infinite canvas, a design agent that reasons across projects, and an agent manager for parallel exploration. Users can generate, iterate, and test full user journeys instantly, including voice-driven design interactions. Stitch also bridges design and development through exports and integrations, allowing teams to move seamlessly from concept to code while maintaining shared design systems across projects.
Importance for marketers: This signals a major shift in creative production. Marketers can rapidly prototype landing pages, campaigns, and user experiences without relying heavily on design or dev teams, accelerating experimentation and reducing time-to-launch.
You can find the previous issue of AI Update here.
Editor’s note: ChatGPT was used to help compile this issue of AI Update.
Enter your email address to continue reading
AI Update, March 20, 2026: AI News and Views From the Past Week
Don’t worry…it’s free!
Sign in with your preferred account, below.
Copy Link
Email
Twitter
Facebook
Pinterest
Linkedin
You may like these other MarketingProfs articles related to Artificial Intelligence:
Keep me signed in
Sign in with your preferred account, below.
Over 350,000 marketers rely on MarketingProfs for B2B know-how every day. Don’t miss out on the latest marketing tips and techniques, delivered right to your inbox.
Subscribe today … it’s free!
Sign in with your preferred account, below.
