Yesterday’s announcements center on one development: agentic commerce moving into general availability. Salesforce released its Agentforce Commerce agents, and the surrounding announcements mark where the gap sits between vendor capability claims and what marketing teams can deploy this quarter.
Salesforce made Shopper Agent, Buyer Agent, and Merchant Agent generally available, with native integration into ChatGPT and Google Search. Two figures define the competitive stakes: AI-referred traffic converts at 8x the rate of social media traffic, and retailers running their own shopper agents grew sales 59% faster during the 2025 holiday season. Those are current-quarter numbers. CMOs who have not deployed owned AI agents on their commerce properties enter the 2026 holiday season without that conversion advantage.
The Conair/Cuisinart case study, published the same day, shows the implementation limits. Amazon’s Creative Agent produced 80% of a video ad concept in four weeks, against the typical three-to-six months. Human editors fixed logos, adjusted brand colors, and corrected errors on the remaining share. The measured gains held (18% higher detail page views, 14% lower cost per detail page view), and the workflow still required human review at each stage. Marketing teams should budget for a hybrid human-AI production model.
Magna5’s shadow AI warning sets a governance cost against the agentic momentum. 49% of employees already use unapproved AI tools at work and share sensitive data with public models. For CMOs holding brand data, customer records, and campaign strategy, this sits inside marketing’s operational domain as a brand-risk and compliance exposure. Magna5 recommends giving employees a sanctioned AI environment, since bans push usage further out of view. The practical choice is whether to govern AI use now or absorb the risk of a data exposure incident later. See AI governance for how leadership structures oversight.
GEO tooling expanded again. XstraStar’s G-Power framework, Visionary Financial’s AI Authority Builder, and the broader set of AI visibility platforms measure brand presence through citation frequency in AI responses, recommendation depth, and competitive positioning in LLM answers. These metrics run alongside rankings and traffic. CMOs face a build-or-partner decision on generative engine optimization capability, and early movers establish reference presence in AI search ahead of competitors.
Splitit’s PYMNTS research defines a constraint for AI-assisted checkout design. 61% of consumers would accept AI-assisted Pay Later recommendations, and 80% of Gen Z would. Only 2% want full automation. Consumers want to review, modify, approve, or reject a recommendation before committing. Credit score protection ranked as the top non-negotiable at 59%, ahead of lowest total cost. AI-assisted checkout flows should reduce friction while keeping a human approval step.
Strategic priorities for marketing leaders
- Deploy owned AI agents on commerce properties ahead of the 2026 holiday season.
- Establish AI governance that gives employees sanctioned tools instead of banning public AI.
- Add GEO metrics to brand measurement dashboards alongside SEO.
- Audit AI-assisted workflows for the hybrid human-AI model that current tools support.
Here’s The News:
Salesforce Unleashes Its Biggest Agentforce Commerce Release Yet — Salesforce made Shopper Agent, Buyer Agent, and Merchant Agent generally available, with native integration into ChatGPT and Google Search (including AI Mode). Gemini app integration follows in coming months. During the 2025 holiday season, AI influenced 20% of global online sales, worth $262 billion, and retailers running their own shopper agents grew sales 59% faster than those without. AI-referred traffic converts at 8x the rate of social media. The Merchant Agent lets merchandisers organize catalogs, sort products, and respond to trends through plain-language commands. Salesforce describes the agents as checking inventory, confirming shipping cutoffs, and closing the sale, beyond the conversational scope of generative chat. These agents transact through emerging agentic commerce protocols. Read the full press release.
Conair Moves Fast to Adopt AI-Driven Video Ads via Amazon’s Creative Agent — Conair Corporation’s e-commerce team published A/B test results for Amazon’s Creative Agent applied to its Cuisinart food processor brand. The AI-generated 15-second video delivered 18% higher detail page views and a 14% reduction in cost per detail page view against a traditionally produced video. Production time dropped from three-to-six months to roughly four weeks. The tool produced 80% of the final video; human editors fixed logos, adjusted brand colors, and corrected errors on the rest. The SVP of e-commerce said the test put the company in an advantageous position in the category. Conair plans to scale the technology and is still defining workflow and prioritization. Read the full article.
Shadow AI Could Expose Sensitive Data Before Companies Know It — Magna5 — Managed IT and cybersecurity provider Magna5 warned that shadow AI carries higher stakes than shadow IT. A recent survey found 49% of workers used AI tools at work without approval, often sharing sensitive data with free versions of tools like ChatGPT. Employees can expose contracts, client records, financial data, and proprietary information through unapproved tools. Magna5 recommends a walled-garden approach: a secure, sanctioned AI environment rather than a ban on public tools, which tends to push usage out of view. The company’s CTO said AI is already inside the business whether leadership approved it or not, and the question is whether companies govern it deliberately or discover the risk after data has left their control. Read the full press release.
AllyHub Launches Browser-Native AI Automation for Ecommerce and Content Teams — AllyHub launched as a browser-native AI copilot that automates execution work across ecommerce, marketing, research, and content production. It operates inside the browser to run research, collect and organize data, generate content, and complete repetitive web workflows. The platform captures successful execution as reusable Skills and Services and measures Return on Token Investment (ROTI). Ecommerce use cases include product and competitor research, marketplace and pricing analysis, product description generation, customer review analysis, and SEO content preparation. Read the full press release.
Splitit Research: 61% of U.S. Consumers Welcome AI-Assisted Pay Later Recommendations, With Control — Splitit released PYMNTS Intelligence research covering 2,034 U.S. consumers. 61% would allow an AI shopping assistant to recommend a Pay Later option for at least one common purchase category, rising to 80% among Gen Z. 2% want full automation; the rest prefer a guided model where they review, modify, approve, or reject a recommendation before committing. Credit score protection ranked as the top non-negotiable at 59%, ahead of lowest total cost. Credit card installments outpaced traditional BNPL three-to-one (36% versus 12% usage in the past three months). Read the full press release.
XstraStar Introduces G-Power Framework for Measuring Brand Presence in Generative AI — XstraStar introduced the G-Power framework, a 0-100 composite score that quantifies a brand’s influence in generative AI responses as part of its GEO methodology. The score weights four dimensions: Visibility (35%), Depth (25%), Recommendation (25%), and Competitiveness (15%). XstraStar notes that appearing in an AI answer does not confirm effective recommendation, since a brand can surface while being presented less favorably than competitors. The framework diagnoses AI visibility gaps and guides content and asset development. Read the full press release.
Visionary Financial Launches AI Authority Builder Platform — VF Digital Assets Media LLC launched AI Authority Builder, combining a free AI website authority audit with a marketplace for PR distribution, editorial placements, and authority-building services. The audit evaluates Website Authority, SEO Visibility, Backlink Profile, Editorial Authority, PR Visibility, Brand Presence, and Authority Growth Opportunities, returning a report in about 60 seconds. The launch reflects rising emphasis on trusted media coverage, authoritative backlinks, and editorial credibility as signals that influence how AI models present brand information. Read the full press release.
Storika Secures Seed Funding to Scale AI-Native Creator Marketing Platform — Storika closed a seed round to scale its AI-native creator marketing platform. The company builds infrastructure that matches brands with creators, optimizes campaign performance, and automates the operational workflows behind influencer programs. The round signals continued investor interest in AI-native approaches to creator marketing. Read the full press release.
S&P Global Evolves Market Intelligence Operating Model to Accelerate Agentic Solutions — S&P Global announced changes to its Market Intelligence operating model to accelerate agentic AI solutions, platform capabilities, and innovation, alongside executive leadership changes. The move shows major financial data providers restructuring around agentic AI delivery. Read the full press release.
airSlate Appoints Former Adobe Executive Claude Alexandre as President — airSlate, the workflow automation and document management platform, appointed former Adobe executive Claude Alexandre as President. The hire brings Adobe enterprise go-to-market experience to a platform expanding its AI-powered document and workflow capabilities. Read the full press release.
Treasure AI Named Leader in 2026 IDC MarketScape for AI-Enabled CDPs — Treasure AI was named a Leader for both B2C and B2B users in the 2026 IDC MarketScape for Worldwide AI-Enabled Customer Data Platforms. The IDC MarketScape now evaluates AI enablement as a core criterion for the CDP category. CMOs evaluating customer data platform investments should weight AI-driven segmentation, prediction, and personalization accordingly. Read the full press release.
MarTech: How to Build a Hermes Agent-Style Workflow in Practice — MarTech published a practical guide to building Hermes Agent-style agentic workflows that reduce AI token costs while holding output quality. Most AI tool workflows send the model more data than the task requires. The Hermes approach uses a local context store, a skill library, and a minimal-prompt extractor so the model receives only the data slice each task needs. Examples show 50-80% token cost reductions for workflows like ad copy generation and social listening. Read the full article.
MarTech: Turning Marketing Complexity into a Competitive Advantage — Infosys SVP John Premkumar published a MarTech analysis arguing that organizations winning with AI have invested in clean data, governance, integrated systems, and team capabilities rather than tool count. Forrester projects worldwide martech spending above $215 billion by 2027, while Gartner data puts martech utilization at 49% industrywide. The piece frames integration, governance, AI depth, and measurement maturity as four pillars of competitive advantage. Read the full article.
Madison Logic Survey: Nearly Half of B2B Marketers Admit They Are Guessing What Drives Purchasing Decisions — A Madison Logic survey with Harris Poll found that nearly half of B2B marketers report guessing what drives purchasing decisions. Despite intent data, AI-powered analytics, and attribution tools, a large share of marketing investment decisions still rest on intuition. The survey points to buyer intelligence infrastructure as the gap. Read the full press release.
