AI marketing news & intelligence — curated daily for marketers

Meta's MTIA 300 Chip: AI Training Gets a Communication Boost

Meta has unveiled its MTIA 300, an in-house AI training chip designed to accelerate recommendation models. Unlike general-purpose GPUs, MTIA 300 integrates Network Interface Controllers (NICs) and communication-offloading engines to significantly improve the speed and efficiency of data transfer between accelerators, a critical bottleneck for complex AI models.

⚡ Meta's custom silicon for AI training highlights a growing trend of specialized hardware optimizing for specific AI workloads, potentially leading to more efficient and cost-effective AI-driven personalization at scale.

Building an AI Content Workflow: Focus on the End Product

Developing a robust AI content workflow requires working backward from the desired final output, defining quality standards before building the system. While AI can significantly boost content production and resource maximization, it carries risks like Google's potential de-indexing of non-commodity content and requires significant upfront investment.

⚡ The true challenge in AI content creation isn't generating text, but engineering a workflow that reliably produces high-quality, publication-ready articles. Brands must weigh the resource gains against the risks of AI-generated content and the significant effort required to build and maintain such systems.

Marketers Still Call AI Search Optimization 'SEO'

Despite the emergence of new acronyms like GEO and AEO, 81% of marketers continue to refer to their AI search visibility strategies as SEO. When seeking help, they are most likely to search for 'AI search optimization' or simply 'SEO'.

⚡ The industry's persistent use of 'SEO' for AI search optimization highlights a disconnect between expert jargon and buyer language, potentially hindering effective vendor selection and strategy development. Marketers are prioritizing clear explanations over buzzwords when evaluating AI marketing solutions.

AI Slop Backlash: SEOs Must Choose Between Shortcuts and Assistance

Platforms like Spotify and LinkedIn are actively combating "AI slop," forcing SEO professionals to re-evaluate their use of artificial intelligence. The core challenge for SEOs is deciding whether to leverage AI as a mere shortcut or as a genuine editorial assistant to enhance their work.

⚡ The backlash against "AI slop" signals a critical juncture for SEOs: relying on AI as a shortcut risks devaluing their expertise and content, while embracing it as an assistant can unlock new levels of efficiency and quality.

Hugging Face Reportedly Valued at $13B Amid Acquisition Talks

AI leader Hugging Face is reportedly in acquisition talks with potential buyers valuing the company at approximately $13 billion. However, the founders' commitment to their community raises questions about whether a sale will ultimately proceed.

⚡ The potential $13B acquisition of Hugging Face underscores the immense market value and strategic importance of open-source AI platforms. Marketers relying on these foundational AI tools should monitor this development closely for potential shifts in access, support, or future direction.

How to use AI personalisation without enterprise-scale resources

General Intuition Raises $6B for AI Robotics with Valor, Point72 Backing

AI startup General Intuition, developing foundation models for generalized AI agents to navigate space and time, is reportedly in talks to raise $6 billion at a pre-money valuation. The funding round is backed by prominent investors including Valor Ventures and Point72 Ventures.

⚡ The significant valuation for General Intuition underscores the immense investor confidence in AI's potential to drive advancements in robotics and physical world interaction, signaling a new frontier for AI applications beyond digital realms.

OpenAI's AI Agents: From Developers to the Masses

OpenAI is actively developing AI agents designed for widespread adoption, aiming to bring this advanced technology beyond software engineers. The article explores the lab's strategy for democratizing access to AI agents.

⚡ OpenAI's push for mass adoption of AI agents signals a significant shift in how users will interact with technology, potentially creating new avenues for personalized marketing and automated customer engagement.

Google Ads Enhances Experimentation with New Search & AI Max Tools

Google Ads has introduced new experimentation tools for both Search and AI Max campaigns. These features allow advertisers to test various settings, including budgets, ROI targets, brand controls, and location targeting, to optimize campaign performance.

⚡ Google's push for AI Max experimentation signals a growing reliance on AI for campaign optimization, forcing advertisers to adapt their testing strategies to leverage these automated insights effectively.

Thomson Reuters Launches Proprietary LLM, 'Thomson', on Data Foundation

Thomson Reuters has launched its own proprietary large language model, 'Thomson', built on its extensive data assets and domain expertise. This in-house model was developed at a significantly lower cost than comparable frontier models by leveraging open-source foundations and a $40 million investment, allowing Thomson Reuters to maintain full ownership and control.

⚡ Thomson Reuters' approach demonstrates that building powerful AI doesn't solely rely on massive compute budgets; a strong data foundation and specialized training can yield cost-effective, controlled frontier models, challenging the 'bigger is better' narrative in AI development.

Wix and Gemini Integrate Website Building into Chat Workflows

Wix has integrated its website building platform directly into Google's Gemini, allowing users to create and manage Wix Harmony websites entirely within the Gemini chat interface. This integration enables users to generate functional websites from natural language prompts and manage business operations and performance metrics without leaving Gemini.

⚡ AI-powered conversational interfaces are blurring the lines between content creation and business operations, enabling rapid deployment of functional websites and business tools directly from chat.

UK ICO guidance on AI in marketing post-Brexit

The UK's Information Commissioner's Office requires marketers using AI for profiling or automated decision-making to conduct a Data Protection Impact Assessment (DPIA). The ICO has signalled it will prioritise audits of AI systems that infer sensitive characteristics from behavioural data.

Google Cloud and Verizon Partner to Scale Enterprise AI

Google Cloud and Verizon have formed a strategic partnership to integrate Google Cloud's full-stack AI, including Gemini Enterprise, into Verizon's operations. This collaboration aims to modernize customer experiences, unify enterprise data, and scale AI across Verizon's business to deliver faster and more intelligent services.

⚡ This partnership signals a significant move by major enterprises to leverage advanced AI for customer experience and operational efficiency, setting a precedent for how telcos will deploy AI to personalize services and manage complex networks.

LoopMe Integrates Brand Signals for Lower-Funnel Performance Optimization

LoopMe's PurchaseLoop solution now uses live brand campaign signals and in-app behavioral data to optimize for lower-funnel goals like app installs, web acquisitions, and ROAS. This expansion aims to address the challenges of post-ATT signal loss and shifting consumer behavior by providing marketers with actionable insights from their own brand campaigns.

⚡ As traditional performance marketing struggles with signal loss, LoopMe's AI-driven approach to leveraging existing brand campaign data for lower-funnel optimization offers a crucial pathway to maintaining effectiveness and ROI.

The Intimacy Economy: AI Shifts Focus from Time-Saving to Meaningful Engagement

The 'Intimacy Economy' redefines marketing goals, emphasizing connection, relevance, and meaning over mere time-saving. AI's role is not to automate tasks, but to free up human capacity for deeper, more impactful interactions with consumers.

⚡ AI is enabling a shift from transactional efficiency to relational depth; marketers must now prioritize building genuine connections and delivering profound value, as AI's true benefit lies in enhancing human engagement, not just automating tasks.

The Hidden Costs of AI Agent Deployments: A TCO Deep Dive

Deploying autonomous AI agents for marketing promises significant efficiency gains, but traditional pricing models fail to capture their true cost. Marketers must look beyond subscription fees to account for hidden expenses like data token usage, custom middleware, and ongoing quality assurance to accurately model the total cost of ownership.

⚡ The allure of agentic AI's efficiency is blinding many to the complex, variable costs that lie beneath the surface. Organizations that fail to model these hidden expenses risk significant budget overruns, undermining the very operational gains they sought.

GoEnhance AI Introduces Visual Localization for Video Production

GoEnhance AI is pioneering 'visual localization,' an AI-driven approach to adapt existing video footage for diverse markets and creative goals. This technology allows brands to alter the artistic presentation of videos while preserving their core performance and structure, offering a more efficient alternative to re-shooting content for different regions or platforms.

⚡ AI-powered visual localization is set to significantly reduce the cost and complexity of global video campaigns, enabling brands to rapidly tailor creative assets to specific audiences without compromising core messaging.

Use AI for keyword research: a 6-step workflow

Findabl.ai Launches AEO Services for AI Search Visibility

Findabl.ai has launched Answer Engine Optimization (AEO) services to help businesses improve their visibility across AI-driven search engines like ChatGPT, Google AI Overviews, and Gemini. This new service focuses on ensuring businesses are clearly represented in the AI-generated answers that are increasingly replacing traditional search results.

⚡ With AI search engines consolidating information into single answers, traditional SEO is no longer enough; businesses must now optimize for AEO to ensure their data is discoverable and accurately represented within these AI-generated responses.

BlueWhale Research Achieves <1% Lead Rejection with ZoomInfo

B2B lead generation firm BlueWhale Research leverages ZoomInfo's AI-powered Go-To-Market platform to achieve a lead rejection rate under 1% for over 45,000 monthly leads. This success stems from using verified contact data and buying signals to target actual researchers, not just delegated executives.

⚡ AI-driven data platforms like ZoomInfo are enabling hyper-accurate B2B targeting, bypassing traditional lead gen inefficiencies and ensuring outreach reaches the true decision-makers, not just the proxies.

Quadient Digital Surges 20%+ in Q2 Driven by AI Automation & E-Invoicing

Quadient Digital reported over 20% year-on-year bookings growth in Q2, propelled by strong demand for its AI-powered business automation and e-invoicing solutions. Key enterprise expansions in financial services and healthcare, alongside significant growth in French e-invoicing ahead of mandates, underscore the company's market momentum.

⚡ The rapid adoption of AI-powered automation and e-invoicing by major enterprises signals a clear market shift towards integrated digital workflows that drive efficiency and compliance. Businesses failing to leverage these AI-driven solutions risk falling behind in operational agility and cost savings.

Demodesk Slashes Prospect Research Time by 90% with ZoomInfo AI

Sales software company Demodesk dramatically reduced daily prospect research time from 2-3 hours to 10-15 minutes by leveraging ZoomInfo's AI GTM platform. This AI-powered solution automates account and contact discovery, freeing up sales reps to focus on selling rather than manual data entry.

⚡ AI-driven sales intelligence platforms like ZoomInfo are proving to be a critical lever for sales efficiency, transforming manual research into a near-instantaneous process and directly impacting quota attainment.

MyDataWork for Teams Now Available on AWS Marketplace

MyDataWork, a platform for organizing analytical assets and AI readiness context, has launched a team edition on AWS Marketplace. This move simplifies procurement for AWS customers, enabling teams to manage dashboards, models, and reports within a shared workspace.

⚡ The availability of MyDataWork for Teams on AWS Marketplace signals a growing need for centralized, context-rich data management solutions that can scale beyond individual use, directly impacting how marketing teams leverage analytics for AI-driven decision-making.

FTC guidelines on AI-generated endorsements and testimonials

The FTC has updated its endorsement guides to cover AI-generated reviews and testimonials. Any AI-written customer quote used in marketing must be clearly labelled as such or risk enforcement action.

AI Training on Books: Copyright Law's Thorny Legal Question

The use of copyrighted books to train AI models without author consent raises significant legal questions, despite the widespread practice. This article explores the complex and evolving legal landscape surrounding AI's reliance on existing creative works.

⚡ AI's reliance on vast datasets, including copyrighted material, is creating a legal minefield that could impact the development and deployment of AI tools. Marketers need to be aware of potential legal challenges and ethical considerations surrounding AI-generated content.

Google's Spam Update, Generative AI in Search, and Reddit's Declining Share

Google's August spam update is now live, coinciding with generative AI features appearing in AI Overviews and a decrease in Reddit's citation share for ChatGPT. These developments signal ongoing shifts in search engine dynamics and AI integration.

⚡ The integration of generative AI into search results, alongside ongoing spam updates, is fundamentally altering how users interact with content and how traffic is distributed. Websites must adapt to these evolving AI-driven search experiences to maintain visibility.

AI Mode Queries Are 3X Longer: Lead With The Answer

Data from AI Mode reveals that users are asking full questions, resulting in queries that are three times longer than traditional search fragments. This shift means that content structured to provide the answer upfront, rather than building to it, will perform better.

⚡ The AI-driven shift to longer, question-based queries demands a content strategy that prioritizes immediate answers. Pages that bury their core message in a narrative arc risk being overlooked by AI assistants and users seeking quick information.

Nvidia: Fine-tuning, Not Model Quality, Drives AI Agent Performance

Nvidia research indicates that the effectiveness of AI agents hinges more on fine-tuning techniques than the inherent quality of the base AI model. This means even less capable AI models can achieve strong performance when properly trained and guided.

⚡ The focus on 'harnessing' AI through fine-tuning, rather than solely on the underlying model's power, suggests that sophisticated AI applications are becoming more accessible. Marketers should prioritize understanding how to effectively guide and refine AI tools for specific tasks, as this skill may be more critical than selecting the most advanced model.