RELVE Q2 Report: 84% of AI News Changed Nothing
RELVE analyzed 1.59 million data points across 23,000+ AI tools and found that 84% of AI news last quarter was "noise" โ real developments with zero operational consequence. Of 182 AI developments tracked, only 29 qualified as actionable "Signals." The three most-covered companies took 28.9% of all coverage yet produced zero signals. The report's central conclusion: the bottleneck is now workflow, governance, and people โ not the technology itself. Marketing and creative tools made up the two largest categories in their shortlist.
Read on Winger Daily โ72% of B2B Buyers Use ChatGPT to Evaluate Vendors โ Most Brands Invisible
Forrester's 2026 B2B Buyer Journey report found 72% of software buyers now consult ChatGPT during vendor evaluation, with 44% using Perplexity specifically during shortlisting. But Crackle PR's Q2 benchmark found 51% of B2B tech brands have zero citations across ChatGPT, Perplexity, and Gemini. The coverage gap is stark: 71% of AI citations come from earned media, not owned content. Press releases fare worst โ only 12% of Google AI Overview answers link to them. For PMMs, the implication is clear: your AI visibility depends more on where you're cited than what you publish.
Read on MarketScale โ95% of B2B Marketers Use AI โ Fewer Than 40% Say It Works
CMI's 2026 B2B research shows universal AI adoption (95%) has not translated to universal results โ only 39% report meaningful performance gains. The insight buried in the data: teams running AI-assisted SDR or content distribution workflows saw a 38% drop in cost-per-lead, per Salesforce. That's fundamentally different from using AI to generate more blog posts. Meanwhile, buying committees have grown to 11.2 stakeholders for deals over $50K, and buyers now review 13.4 pieces of content before contacting a vendor. Most of that research happens without any direct vendor interaction.
Read on MarketScale โChina Implements World's First AI Agent Regulations
China rolled out binding rules for AI agents this week โ the "Implementation Opinions on the Standardized Application and Innovative Development of Intelligent Agents" โ establishing a tiered decision-authorization system for agent autonomy and data access. Separate measures govern "emotionally interactive" agents, banning minors from virtual companion services, requiring intervention for emotional dependence, and forbidding use of private conversations for training. At WAIC 2026, Alibaba showcased Qwen Office with multi-agent orchestration, signaling a pivot from model races to deployed agent systems. For PMMs watching the regulatory landscape: these rules preview how governments may treat autonomous and emotionally engaging agents differently from chatbots.
Read on AI Agent Store โ"Batch-and-Blast" Email is Dead โ Agentic Workflows Take Over
Salesfully declares the traditional B2B email playbook officially dead. Stricter inbox deliverability rules and AI-powered email triage on the recipient side mean generic marketing blasts get filtered before humans see them. The replacement: autonomous agentic workflows that analyze live company triggers โ funding rounds, executive hires, tech stack changes โ and dynamically craft context-aware outreach in real-time. The shift demands moving from broadcast marketing to hyper-personalized, one-to-one engagement at scale. Mass templates with {{FirstName}} tags are no longer competitive.
Read on Salesfully โ๐ก My Take
Read this one: The RELVE report on signal vs. noise. When 84% of AI news changes nothing about how you build, staff, budget, or buy โ and the most-covered companies produce zero actionable signals โ it's a reminder that attention and consequence point in opposite directions. The bottleneck isn't technology anymore. It's knowing what to do with it.